The DOI matches its Crossref record; 142 of the 142 references it cites were found in a registry or an index; none of the 144 sources its claims cite is checked for support yet.
The Fallacy of AI Functionality
Open accessAbstract
Deployed AI systems often do not work. They can be constructed haphazardly, deployed indiscriminately, and promoted deceptively. However, despite this reality, scholars, the press, and policymakers pay too little attention to functionality. This leads to technical and policy solutions focused on “ethical” or value-aligned deployments, often skipping over the prior question of whether a given system functions, or provides any benefits at all. To describe the harms of various types of functionality failures, we analyze a set of case studies to create a taxonomy of known AI functionality issues. We then point to policy and organizational responses that are often overlooked and become more readily available once functionality is drawn into focus. We argue that functionality is a meaningful AI policy challenge, operating as a necessary first step towards protecting affected communities from algorithmic harm.
Overview
AI snapshot
Written from the paper’s full text.
TL;DR
The paper argues that AI policy and ethics discourse often assumes deployed AI systems function, and that this assumption obscures a broad class of harms from dysfunctional or non-working systems. It builds a taxonomy of AI functionality failures from 283 cases and argues functionality should be treated as a primary policy and legal concern.
Key findings
- The “functionality assumption” appears across AI ethics guidelines, trust framing, democratization efforts, and debates about misuse or autonomous agents; critiques often engage in “criti-hype” and skip the prior question of whether systems work.
- From 283 cases (2012–2021), the paper develops a failure taxonomy: impossible tasks (conceptually and practically impossible), engineering failures (design, implementation, missing safety features), deployment failures (robustness, adversarial attacks, unanticipated interactions), and communication failures (falsified/overstated and misrepresented capabilities).
- Examples include Michigan’s MIDAS algorithm falsely flagging over 20,000 people for unemployment fraud; ETS voice recognition identifying 97% of UK tests as “suspicious,” classifying 58% of 58,459 tests as “invalid” and 39% as “questionable,” with disputed error rates from 1% to 30% and only 3,600 of 12,500 appeals won; Amazon Rekognition falsely matching 28 members of Congress; IBM Watson at MD Anderson costing $62 million with no results; and Indiana welfare automation producing a 54% increase in denials.
- The paper points to overlooked responses: FTC Section 5 unfair/deceptive practices authority, products liability, warranties, fraud, internal audits/documentation, product certification and standards, procurement requirements, safety culture, and monitoring tools.
Methods
- Systematic review of case studies using the AI, Algorithmic and Automation Incident and Controversy Repository (AIAAIC) spreadsheet; from over 800 cases, filtered to 283 cases from 2012 to 2021 based on whether the technology claimed to be AI/ML/data-driven and whether harm was due to a technology failure, focusing on artifact-connected failures rather than infrastructural or environmental “meta” failures.
- Iterative tagging procedure to develop the taxonomy; tags were updated, merged, and grouped in meetings between tagging sessions, then known media and academic case studies were selected to illustrate failure modes.
- Qualitative taxonomy-building and legal/policy analysis; no quantitative model or experiment is reported.
Limitations
- The paper states functionality is difficult to define precisely: “fitness for a product’s intended use” is incomplete because some intended uses are impossible, vendor performance claims may be insufficient, and “meeting stakeholder expectations” is too broad.
- The methodology relies partly on AIAAIC, a crowdsourced repository from journalism professionals; filtering and tagging are interpretive, and the analysis focuses on artifact-connected failures while excluding infrastructural or environmental “meta” failures.
- The paper states self-regulatory organizational interventions are far from adequate oversight due to conflicts of interest, and internal audits rarely communicate results externally, with disclosure neither mandatory nor incentivized.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
87 Sources not found · 84 Own interpretation · 19 No basis given · 3 Not checked yet · 1 Own result
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“We then argue that this assumption does not hold in a large set of cases.”
§ The Fallacy of AI Functionality > 1 INTRODUCTION
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
None of the sources cited for this claim have been checked yet.
In the paper
“Deployed AI systems often do not work.”
- Also stated in: § 1 INTRODUCTION · plain
“Despite the current public fervor over the great potential of AI, many deployed algorithmic products do not work.”
§ The Fallacy of AI Functionality > Abstract
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Also stated in: § 1 INTRODUCTION · plain
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“This highly opaque system was thus notable for the disproportionate amount of emphasis that was put into finding cheaters rather than protecting those who were falsely accused.”
§ The Fallacy of AI Functionality > Model Design Failures.
plainn = 12,500- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Failures in AI systems can often be traced to these specific policies or decisions in the development process of the system.”
§ The Fallacy of AI Functionality > 4.2.2 Engineering Failures.
plainconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
None of the sources cited for this claim could be matched to a paper in Crossref, so there is nothing to check it against.
In the paper
“For example, in the absence of any reasonable observable characteristics or accessible data to measure the model goals in question, attempts to represent these objectives end up being inappropriate proxies.”
§ The Fallacy of AI Functionality > Practically Impossible.
What the sources say
- Not matchedcites [98]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites [99]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Not matched
None of the sources cited for this claim could be matched to a paper in Crossref, so there is nothing to check it against.
In the paper
“These choices are normative and may prioritize values such as efficiency over preventing harmful failures [47,117].”
§ The Fallacy of AI Functionality > Model Design Failures.
What the sources say
- Not matchedcites [47]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites [117]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: unclear — It is unclear whether it could be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Not matched
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“In this setting, there is no logical basis with which to claim functionality.”
§ The Fallacy of AI Functionality > Conceptually Impossible.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: unclear — It is unclear whether it could be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
In the paper
“One area in which a lack of sufficient testing has been observed in the development of AI is in the area of clinical medicine.”
§ The Fallacy of AI Functionality > Model Implementation Failures.
What the sources say
- Not matchedcites [129]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Not matched
None of the sources cited for this claim could be matched to a paper in Crossref, so there is nothing to check it against.
In the paper
“Fairness research is often centered around attempting to balance some notion of accuracy with some notion of fairness [59,63,68].”
§ The Fallacy of AI Functionality > 3 THE FUNCTIONALITY ASSUMPTION
What the sources say
- Not matchedcites [59]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites [63]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites [68]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Not matched
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Sometimes, the design specifications of a model are inappropriate for the task it is being developed for.”
§ The Fallacy of AI Functionality > Model Design Failures.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Probably not a substantial claim. 74
- It may state two separate claims as one. 65
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Firstly, the sheer number of points of failure we were able to identify illustrates the scope of the problem.”
§ The Fallacy of AI Functionality > 4 THE MANY DIMENSIONS OF AI DYSFUNCTION
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim.
- Rests on an assumption: The iterative tagging procedure reliably and validly categorizes each case into the failure modes of the taxonomy.
None of the sources cited for this claim could be matched to a paper in Crossref, so there is nothing to check it against.
In the paper
“For example, audits of AI hiring tools focus primarily on ensuring an 80% selection rate for protected classes (the so-called 4/5ths rule) is satisfied, and rarely mention product validation processes, demonstrating an assumed validity of the prediction task [52,148,199].”
§ The Fallacy of AI Functionality > 3 THE FUNCTIONALITY ASSUMPTION
What the sources say
- Not matchedcites [52]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites [148]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites [199]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
boosteddirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Not matched
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“As an overlooked aspect of AI policy, functionality is often presented as a consideration secondary to other ethical challenges.In this paper, we argue that it is a primary concern that often precedes such problems.”
§ The Fallacy of AI Functionality > 1 INTRODUCTION
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: unclear — It is unclear why the paper says it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“However, despite this reality, scholars, the press, and policymakers pay too little attention to functionality.”
§ The Fallacy of AI Functionality > Abstract
plain- Asserted: passes — The authors put it forward as true.
- Checkable: fails — It may be too vague to check against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence. The paper may give no reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“We start by calling out what we perceive to be a functionality assumption, prevalent in much of the discourse on AI risks.”
§ The Fallacy of AI Functionality > 1 INTRODUCTION
hedged- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper may give no reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“It is unsurprising that promoters of AI do not tend to question its functionality.”
§ The Fallacy of AI Functionality > 3 THE FUNCTIONALITY ASSUMPTION
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Similarly, to the extent regulators are looking into functionality, explanations aimed at regulators can assist oversight, but typically explanations are desired to check the basis for decisions, while assuming the systems work as intended.”
§ The Fallacy of AI Functionality > 3 THE FUNCTIONALITY ASSUMPTION
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“It is possible for designers to preempt misuse by implementing instructions, warnings, or error conditions, and failure to do so creates a system that does not function properly.”
§ The Fallacy of AI Functionality > Unanticipated Interactions.
hedgeddirect- Asserted: passes — The authors put it forward as true.
- Checkable: fails — It may be too vague to check against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence. The paper may give no reason why it holds. It may state two separate claims as one.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“The accuracy of ETS's method was disputed between experts sought by the National Union of Students and the Home Office; the resulting estimates of error rates ranged from 1% to 30%.”
§ The Fallacy of AI Functionality > Model Design Failures.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
In the paper
“Under the hood, however, Instead of software producing financial statements, dozens of accountants did most of it manually from ScaleFactor's Austin headquarters or from an outsourcing office in the Philippines, according to former employees.”
§ The Fallacy of AI Functionality > 4.2.4 Communication Failures.
What the sources say
- Not matchedcites [100]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Not matched
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Due to upstream policy, data or societal choices, AI tasks can be practically impossible for one set of developers and not for another, or for different reasons in different contexts.”
§ The Fallacy of AI Functionality > Practically Impossible.
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“We then discuss the existing accountability tools to address functionality issues, that are often overlooked in AI policy literature and in practice, due in large part to this assumption of functionality.”
§ The Fallacy of AI Functionality > 1 INTRODUCTION
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper may give no reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“While this in and of itself may not create unsafe products, expectations based on unreasonable claims can create unearned trust, and a potential overreliance that hurts parties who purchase the product.”
§ The Fallacy of AI Functionality > 4.2.4 Communication Failures.
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“This research question presumes that an unconstrained solution without fairness restrictions is the optimal solution to the problem.”
§ The Fallacy of AI Functionality > 3 THE FUNCTIONALITY ASSUMPTION
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Functionality can be difficult to define precisely.”
§ The Fallacy of AI Functionality > 4 THE MANY DIMENSIONS OF AI DYSFUNCTION
plain- Asserted: passes — The authors put it forward as true.
- Checkable: fails — It may be too vague to check against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence.
In the paper
“Similarly, Raji et al. [149] argue that AI makes claims to generality while modeling behaviour that is determined by highly constrained and context-specific data.”
§ The Fallacy of AI Functionality > 2 RELATED WORK
What the sources say
- Not matchedcites [149]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
- Not matched
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Sometimes model failures are anticipated yet difficult to prevent; in this case, engineers can sometimes take steps to ensure these points of failure will not cause harm.”
§ The Fallacy of AI Functionality > Missing Safety Features.
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
None of the sources cited for this claim could be matched to a paper in Crossref, so there is nothing to check it against.
In the paper
“Other papers discuss how the AI label lends itself to inflated claims of functionality that the systems cannot meet.”
§ The Fallacy of AI Functionality > 2 RELATED WORK
What the sources say
- Not matchedcites [102]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites [28]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim.
- Not matched
In the paper
“Of the 81 non-randomized studies, they found the median number of clinical experts compared to the AI was 4, full access to datasets and code were unavailable in over 90% of studies, the overall risk of bias was high, and adherence to reporting standards were suboptimal, and therefore poorly substantiate their claims.”
§ The Fallacy of AI Functionality > Model Implementation Failures.
What the sources say
- Not matchedcites [129]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Not matched
In the paper
“Similarly, the Epic sepsis prediction model, a product actually implemented at hundreds of hospitals, was recently externally validated by Wong et al. [203], who found that the model had poor calibration to other hospital settings and discriminated against under-represented demographics.”
§ The Fallacy of AI Functionality > Model Implementation Failures.
What the sources say
- Not matchedcites [203]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Not matched
In the paper
“There is also some work demonstrating that AI products are challenging to engineer correctly in practice.”
§ The Fallacy of AI Functionality > 2 RELATED WORK
What the sources say
- Not matchedcites [194]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
boosted- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
- Not matched
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“Yet out of 12,500 people who appealed their immigration decisions, only 3,600 won their cases-and only a fraction of these were won through actually disproving the allegations of cheating.”
§ The Fallacy of AI Functionality > Model Design Failures.
plainn = 12,500- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: unclear — It is unclear why the paper says it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Communication failures often co-occur with other functional safety problems, and the lack of accountability for false claims -intentional or otherwise -makes these particularly pernicious and likely to occur as AI hype continues absent effective regulation.”
§ The Fallacy of AI Functionality > 4.2.4 Communication Failures.
hedgeddirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
- Rests on an assumption: AI hype will continue absent effective regulation.
In the paper
“Similarly, as an early example of technical fairness solutions, Feldman et al. [60] created a method to remove disparate impact from a model while preserving rank, which only makes sense if the unconstrained system output is correct in the first place.”
§ The Fallacy of AI Functionality > 3 THE FUNCTIONALITY ASSUMPTION
What the sources say
- Not matchedcites [60]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plainconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Not matched
In the paper
“Some customers say they received books filled with errors, and were forced to re-hire accountants, or clean up the mess themselves.[100]”
§ The Fallacy of AI Functionality > 4.2.4 Communication Failures.
What the sources say
- Not matchedcites [100]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
hedgeddirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Not matched
None of the sources cited for this claim could be matched to a paper in Crossref, so there is nothing to check it against.
In the paper
“As an example, he points to disinformation researchers, who embrace corporate talking points of a recommendation model that can meaningfully influence consumer behavior to the point of controlling their purchases or voting activity-when in actuality, these algorithms have little ability to do either [22,75,88,95,162].”
§ The Fallacy of AI Functionality > 2 RELATED WORK
What the sources say
- Not matchedcites [22]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites [75]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites [88]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites [95]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites [162]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
- Not matched
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“These proposals thus often fall short of acknowledging the functionality issues in AI deployments and the role of the lack of functional safety in contributing to the harm perpetuated by these systems.”
§ The Fallacy of AI Functionality > 1 INTRODUCTION
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim.
In the paper
“From the engineering side, interpretability is usually desired for debugging purposes [23], so it is focused on functionality.”
§ The Fallacy of AI Functionality > 3 THE FUNCTIONALITY ASSUMPTION
What the sources say
- Not matchedcites [23]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Not matched
None of the sources cited for this claim could be matched to a paper in Crossref, so there is nothing to check it against.
In the paper
“Recent work details that the AI research field suffers from scientific validity and evaluation problems [48,79].”
§ The Fallacy of AI Functionality > 2 RELATED WORK
What the sources say
- Not matchedcites [48]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites [79]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
- Not matched
In the paper
“As with other areas of software development, roles in AI development and deployment are becoming more specialized.”
§ The Fallacy of AI Functionality > 4.2.4 Communication Failures.
What the sources say
- Not matchedcites [44]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
- Not matched
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“These mentions suggest that functionality is not always ignored outright.Sometimes, it is considered in policy, but in many cases, that consideration lacks the emphasis of the other concerns presented.”
§ The Fallacy of AI Functionality > 3 THE FUNCTIONALITY ASSUMPTION
hedged- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Many of these cases are direct refutations of the specific instances of the functionality assumptions in Section 3.”
§ The Fallacy of AI Functionality > 4.2 Failure Taxonomy
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim.
- Rests on an assumption: The iterative tagging procedure reliably and validly categorizes each case into the failure modes of the taxonomy.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Functionality could also be seen as a statement that a product lives up to the vendor's performance claims, but this, too, is incomplete; specifications chosen by the vendor could be insufficient to solve the problem at hand.”
§ The Fallacy of AI Functionality > 4 THE MANY DIMENSIONS OF AI DYSFUNCTION
hedged- Asserted: passes — The authors put it forward as true.
- Checkable: fails — It may be too vague to check against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence. The paper’s conclusions may not depend on it. It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“For instance, in a classification model, choices such as which input and target variables to use, whether to prioritize accepting true positives or rejecting false negatives, and how to process the training data all factor into determining model outcomes.”
§ The Fallacy of AI Functionality > Model Design Failures.
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Communication failures often co-occur with other functional safety problems, and the lack of accountability for false claims -intentional or otherwise -makes these particularly pernicious and likely to occur as AI hype continues absent effective regulation.”
§ The Fallacy of AI Functionality > 4.2.4 Communication Failures.
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds.
None of the sources cited for this claim could be matched to a paper in Crossref, so there is nothing to check it against.
In the paper
“While these are credible threats, concerns around large language models tend to dismiss the practical limitations of what these models can achieve [18], neglecting to address more mundane hazards tied to the premature deployment of a system that does not work [55,189].”
§ The Fallacy of AI Functionality > 3 THE FUNCTIONALITY ASSUMPTION
What the sources say
- Not matchedcites [18]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites [55]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites [189]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Not matched
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Data validation is a critical step in the construction of a ML system, and the team that built the benefit system chose to use a highly problematic dataset to train their model.”
§ The Fallacy of AI Functionality > Model Implementation Failures.
plaindirectSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
None of the sources cited for this claim could be matched to a paper in Crossref, so there is nothing to check it against.
In the paper
“While it is not controversial to claim that such concerns are far from being realized [13,42,146], this fear of misspecified objectives, runaway feedback loops, and AI alignment presumes the existence of an industry that can get AI systems to execute on any clearly declared objectives, and that the main challenge is to choose and design an appropriate goal.”
§ The Fallacy of AI Functionality > 3 THE FUNCTIONALITY ASSUMPTION
What the sources say
- Not matchedcites [13]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites [42]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites [146]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Not matched
In the paper
“However, the lack of a failsafe to make sure the carbon monoxide detector could not be turned off accidentally made the product dangerous.”
§ The Fallacy of AI Functionality > Missing Safety Features.
What the sources say
- Not matchedcites [200]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plaindirectSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Not matched
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“We argue that functionality is a meaningful AI policy challenge, operating as a necessary first step towards protecting affected communities from algorithmic harm.”
§ The Fallacy of AI Functionality > Abstract
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: unclear — It is unclear whether it could be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
None of the sources cited for this claim could be matched to a paper in Crossref, so there is nothing to check it against.
In the paper
“For example, when large language models are used to process the queries of AI-powered web search [131], the models' fragility to misspellings [125,147], or trivial changes to format [19] and context [18] can lead to unexpected results.”
§ The Fallacy of AI Functionality > Robustness Issues.
What the sources say
- Not matchedcites [131]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites [125]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites [147]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites [19]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Not matched
In the paper
“the data they used to come up with their formula for setting people's assistance limits was corrupt.”
- Also stated in: § Model Implementation Failures. · plain
“they had to throw out two-thirds of the records they had before they came up with the formula because of data entry errors and data that didn't make sense.”
§ The Fallacy of AI Functionality > Model Implementation Failures.
What the sources say
- Not matchedcites [177]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Also stated in: § Model Implementation Failures. · plain
In the paper
“Some roles focus on managing the data that feeds into models, others specialize in modeling, and others optimally engineer models for speed and scale [44].”
§ The Fallacy of AI Functionality > 4.2.4 Communication Failures.
What the sources say
- Not matchedcites [44]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
- Not matched
In the paper
“The draft AI Act in the EU includes requirements for pre-marketing controls to establish products' safety and performance, as well as quality management for high risk systems [190].”
§ The Fallacy of AI Functionality > 3 THE FUNCTIONALITY ASSUMPTION
What the sources say
- Not matchedcites [190]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Not matched
In the paper
“Krafft et al. [110] found that policymakers often define AI with respect to how human-like a system is, and concluded that this could lead to deprioritizing issues more grounded in reality.”
§ The Fallacy of AI Functionality > 2 RELATED WORK
What the sources say
- Not matchedcites [110]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
- Not matched
None of the sources cited for this claim could be matched to a paper in Crossref, so there is nothing to check it against.
In the paper
“US EHR data is often erroneous, miscoded, fragmented, and incomplete [91,92], creating a mismatch between available data and intended use [74].”
§ The Fallacy of AI Functionality > Practically Impossible.
What the sources say
- Not matchedcites [91]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites [92]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites [74]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Not matched
None of the sources cited for this claim could be matched to a paper in Crossref, so there is nothing to check it against.
In the paper
“For example, expressed concerns around large language models centers on hyped narratives of the models' ability to generate hyper-realistic online content, which could theoretically be used by malicious actors to facilitate harmful misinformation campaigns [176,195].”
§ The Fallacy of AI Functionality > 3 THE FUNCTIONALITY ASSUMPTION
What the sources say
- Not matchedcites [176]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites [195]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim.
- Not matched
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The hype surrounding IBM's Watson in healthcare represents another example where a product that may have been fully capable of performing specific helpful tasks was sold as a panacea to health care's ills.”
§ The Fallacy of AI Functionality > Misrepresented Capabilities.
hedged- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“Sales of goods typically come with a set of implied warranties.”
§ The Fallacy of AI Functionality > 5.1.3 Warranties.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Thankfully, as AI operates as a general purpose technology prevalent in many of these industries, there already exists a plethora of governance infrastructure to address this issue in related fields of application.”
§ The Fallacy of AI Functionality > 5 DEALING WITH DYSFUNCTION: OPPORTUNITIES FOR INTERVENTION ON FUNCTIONAL SAFETY
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds.
- Rests on an assumption: AI operates as a general purpose technology prevalent in many industries.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“In addition to legal levers, there are many organizational interventions that can be deployed to address the range of functionality issues discussed.”
§ The Fallacy of AI Functionality > 5.2 Organizational interventions
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: fails — It may be too vague to check against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence. The paper’s conclusions may not depend on it. The paper may give no reason why it holds.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“In general, if a person is injured by a defective product, they can sue the producer or seller in products liability.”
§ The Fallacy of AI Functionality > 5.1.2 Products Liability Law.
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“There are several other organizational factors that can determine and assess the functional safety of a system.”
§ The Fallacy of AI Functionality > 5.2.3 Other Interventions.
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Existing functional safety standards from the automobile (ISO 26262), aerospace (US RTCA DO-178C), defense (MIL-STD-882E) and electronics (IEEE IEC 61508 / IEC 61511) industries, amongst others, can provide a template on how to approach this challenge within the AI industry.”
§ The Fallacy of AI Functionality > 5.2.2 Product Certification & Standards.
plainconditional- Asserted: passes — The authors put it forward as true.
- Checkable: fails — It may be too vague to check against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence. The paper’s conclusions may not depend on it. The paper may give no reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The AI field seems to now be facing this same challenge.”
§ The Fallacy of AI Functionality > 5 DEALING WITH DYSFUNCTION: OPPORTUNITIES FOR INTERVENTION ON FUNCTIONAL SAFETY
hedged- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Products liability could especially easily apply to engineering failures, where the error was foreseeable and an alternative, working version of the product should have been built.”
§ The Fallacy of AI Functionality > 5.1.2 Products Liability Law.
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“Fraud has many specific meanings in law, but invariably it involves a knowing or intentional misrepresentation that the victim relied on in good faith.”
§ The Fallacy of AI Functionality > 5.1.4 Fraud.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Defunct AI will breach both these warranties.”
§ The Fallacy of AI Functionality > 5.1.3 Warranties.
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: unclear — It is unclear why the paper says it holds.
It may state two separate claims as one.
In the paper
“CEN (European Committee for Standardisation) and CENELEC (European Committee for Electrotechnical Standardisation), two of three European Standardisation Organisations (ESOs) were heavily involved in the creation of the EU's draft AI Act [190].”
§ The Fallacy of AI Functionality > 5.2.2 Product Certification & Standards.
What the sources say
- Not matchedcites [190]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
boostedcorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
- Not matched
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“We cannot take for granted that AI products work.”
§ The Fallacy of AI Functionality > 6 CONCLUSION : THE ROAD AHEAD
plain- Asserted: passes — The authors put it forward as true.
- Checkable: fails — It may be too vague to check against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence.
In the paper
“The AI field could certainly adopt similar techniques for increasing the scrutiny of their systems, especially given the nascent state of reflection and standardization common in ML evaluation processes [118].”
§ The Fallacy of AI Functionality > 5.2.1 Internal Audits & Documentation.
What the sources say
- Not matchedcites [118]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: fails — It may be too vague to check against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence. The paper’s conclusions may not depend on it.
- Not matched
In the paper
“Similarly, Eubanks [56] demonstrates how a $1.4 billion contract with a coalition of hightech companies led an Indiana governor to stubbornly continue a welfare automation algorithm that resulted in a 54% increase in the denials of welfare applications.”
§ The Fallacy of AI Functionality > Misrepresented Capabilities.
What the sources say
- Not matchedcites [56]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
boosteddirectSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Not matched
In the paper
“Choi analogizes this issue to car crashes: car crashes are inevitable, but courts developed the idea of crashworthiness to ask about the car's contribution to the total harm, even if the initial injury was attributable to a product defect [36].”
§ The Fallacy of AI Functionality > 5.1.2 Products Liability Law.
What the sources say
- Not matchedcites [36]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may state two separate claims as one.
- Not matched
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“It would stand to reason that a functionality failure in an AI system could be deemed a product defect.”
§ The Fallacy of AI Functionality > 5.1.2 Products Liability Law.
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“In extreme cases, the sale of defunct AI may constitute fraud.”
§ The Fallacy of AI Functionality > 5.1.4 Fraud.
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Thus, where dysfunctional AI is being sold and its failures causes substantial harm to consumers, the FTC could step in.”
§ The Fallacy of AI Functionality > 5.1.1 Consumer Protection.
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: fails — It may be too vague to check against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“The National Highway Safety Administration offers similar oversight for automobiles specifically.”
§ The Fallacy of AI Functionality > 5.1.1 Consumer Protection.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“If a system is not functional, in the sense we describe, a court should have no problem finding that it is unreasonably defective.”
§ The Fallacy of AI Functionality > 5.1.2 Products Liability Law.
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: fails — It may be too vague to check against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The fact that faulty AI products are on the market today makes this problem particularly urgent.”
§ The Fallacy of AI Functionality > 6 CONCLUSION : THE ROAD AHEAD
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper may give no reason why it holds. It may state two separate claims as one.
In the paper
“In addition to various federal agencies, all states have consumer protection statutes that bar deceptive practices and many bar unfair practices as well, like the FTC Act [33].”
§ The Fallacy of AI Functionality > 5.1.1 Consumer Protection.
What the sources say
- Not matchedcites [33]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Not matched
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Another area of law implicated by product failure is warranty law, which protects the purchasers of defunct AI and certain third parties who stand to benefit from the sale.”
§ The Fallacy of AI Functionality > 5.1.3 Warranties.
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
In the paper
“State attorneys general often take active roles as enforcers of those laws [38].”
§ The Fallacy of AI Functionality > 5.1.1 Consumer Protection.
What the sources say
- Not matchedcites [38]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
- Not matched
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“This follows a pattern seen with machine learning products in general, where the advertised performance on a simple and static data benchmark, is much lower than the performance on the often more complex and diverse data encountered in practice.”
§ The Fallacy of AI Functionality > 4.2.4 Communication Failures.
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“This would likely not be a problem for software that resides within a physical system, but for non-embodied AI, it might pose a hurdle.”
§ The Fallacy of AI Functionality > 5.1.2 Products Liability Law.
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
In the paper
“Similarly, cultural expectations for safety and engineering responsibility impact the quality of the output from the product development process -setting these expectations internally and fostering a healthy safety culture can increase cooperation on other industrywide and organizational measures [163].”
§ The Fallacy of AI Functionality > 5.2.3 Other Interventions.
What the sources say
- Not matchedcites [163]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Not matched
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“Injuries not related to the breach of contract are meant to be worked out in tort law, as described above.”
§ The Fallacy of AI Functionality > 5.1.3 Warranties.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“This is particularly a problem in the context of algorithms developed by public agencies -where the AI systems can be adopted as symbols for progress, or smokescreens for undesirable policy outcomes, and thus liable to inflated narratives of performance.”
§ The Fallacy of AI Functionality > Misrepresented Capabilities.
plainconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“The Commission does not have general rulemaking authority, so most of its actions to date have taken the form of public education and enforcement.”
§ The Fallacy of AI Functionality > 5.1.1 Consumer Protection.
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
In the paper
“Taking the form of anything from documentation exercises to challenge datasets as benchmarks, these processes raised the bar for deployment criteria and matured the product development pipeline in the process [152].”
§ The Fallacy of AI Functionality > 5.2.1 Internal Audits & Documentation.
What the sources say
- Not matchedcites [152]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Not matched
None of the sources cited for this claim could be matched to a paper in Crossref, so there is nothing to check it against.
In the paper
“Building on previous work [31], the ACLU ran Rekognition with a database of mugshots against members of U.S. Congress using the default setting and found 28 members falsely matched within the database, with people of color shown as a disproportionate share of these errors [175].”
§ The Fallacy of AI Functionality > Misrepresented Capabilities.
What the sources say
- Not matchedcites [31]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites [175]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Not matched
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“A trickier intervention is the avenue of product certification and standards development for AI products.”
§ The Fallacy of AI Functionality > 5.2.2 Product Certification & Standards.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: fails — It may be too vague to check against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence. The paper’s conclusions may not depend on it.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“As discussed above, a product could be placed on the market claiming the ability to do something it cannot achieve in theory or in practice, or it can fail to be robust to unanticipated but foreseeable uses by consumers.”
§ The Fallacy of AI Functionality > 5.1.2 Products Liability Law.
hedged- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“The Commission does, however, have authority to make rules regarding unfair or deceptive practices under the Magnuson-Moss Warranty Act.”
§ The Fallacy of AI Functionality > 5.1.1 Consumer Protection.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“In many cases, it required the active advocacy of concerned citizens to lead to the policy interventions that would effectively change the tide of these industries.”
§ The Fallacy of AI Functionality > 5 DEALING WITH DYSFUNCTION: OPPORTUNITIES FOR INTERVENTION ON FUNCTIONAL SAFETY
hedgeddirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“False advertising laws are related and also common.”
§ The Fallacy of AI Functionality > 5.1.1 Consumer Protection.
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Poorly vetted products permeate our lives, and while many readily accept the potential for harms as a tradeoff, the claims of the products' benefits go unchallenged.”
§ The Fallacy of AI Functionality > 6 CONCLUSION : THE ROAD AHEAD
plain- Asserted: passes — The authors put it forward as true.
- Checkable: fails — It may be too vague to check against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence. The paper may give no reason why it holds. It may state two separate claims as one.
- Rests on an assumption: Poorly vetted AI products permeate our lives.
- Rests on an assumption: The claims of AI products' benefits go unchallenged.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Although functionality issues are not completely ignored in AI policy, the lack of awareness of the range in which these issues arise leads to the problems being inadequately emphasized and poorly addressed by the full scope of accountability tools available.”
§ The Fallacy of AI Functionality > 6 CONCLUSION : THE ROAD AHEAD
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper may give no reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Another avenue for legal accountability may come from the tort of products liability, though there are some potential hurdles.”
§ The Fallacy of AI Functionality > 5.1.2 Products Liability Law.
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
In the paper
“Though it has created no new rules since 1980, in July 2021, the FTC voted to change internal agency policies to make it easier to do so [41].”
§ The Fallacy of AI Functionality > 5.1.1 Consumer Protection.
What the sources say
- Not matchedcites [41]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Not matched
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Fraud is difficult to prove, and our examples do not clearly indicate fraud, but it is theoretically possible if someone is selling snake oil.”
§ The Fallacy of AI Functionality > 5.1.4 Fraud.
hedged- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“As a result, assessment outcomes are mainly for internal use only, often just to set internal quality assurance standards for deployment and prompt further engineering reflection during the evaluation process.”
§ The Fallacy of AI Functionality > 5.2.1 Internal Audits & Documentation.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
In the paper
“Due process requires decisions to not be arbitrary, and AI that doesn't work loses its claim to making decisions on a sound basis [37].”
§ The Fallacy of AI Functionality > 5.1.5 Other Legal Avenues Already Being Explored.
What the sources say
- Not matchedcites [37]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Not matched
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“Fraud can lead to criminal liability as well.”
§ The Fallacy of AI Functionality > 5.1.4 Fraud.
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds.
In the paper
“At Memorial-Sloan Kettering, after $62 million dollars spent and may years of effort, MD Anderson famously cancelled IBM Watson contracts with no results to show for it [89].”
§ The Fallacy of AI Functionality > Misrepresented Capabilities.
What the sources say
- Not matchedcites [89]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
boosted- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Not matched
None of the sources cited for this claim could be matched to a paper in Crossref, so there is nothing to check it against.
In the paper
“However, the study used to make this claim (still prominently displayed in press release materials) was quickly debunked by a series of outside researchers who found that at the document-level, when provided with context from nearby sentences, and/or compared to human experts, the machine translation model did not indeed achieve equal accuracy to human translators [114,185].”
§ The Fallacy of AI Functionality > 4.2.4 Communication Failures.
What the sources say
- Not matchedcites [186]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites [114]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites [185]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Not matched
None of the sources cited for this claim could be matched to a paper in Crossref, so there is nothing to check it against.
In the paper
“Also, as functionality is a safety risk aligned with profit-oriented goals, many model logging and evaluation operations tools are available for organizations to leverage in the internal inspection of their systems -including tools for more continuous monitoring of deployed systems [154,169].”
§ The Fallacy of AI Functionality > 5.2.3 Other Interventions.
What the sources say
- Not matchedcites [154]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites [169]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
- Rests on an assumption: Functionality is a safety risk aligned with profit-oriented goals.
- Not matched
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“In tort law, a claim of fraud can lead to compensation necessary to rectify any harms that come from the fraud, as well as punitive damages in egregious cases.”
§ The Fallacy of AI Functionality > 5.1.4 Fraud.
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
In the paper
“The implied warranty of fitness for particular purpose applies when a seller knows that the buyer has a specific purpose in mind and the buyer is relying on the seller's skill or judgment about the good's fitness, stating that the good is fit for that purpose [188].”
§ The Fallacy of AI Functionality > 5.1.3 Warranties.
What the sources say
- Not matchedcites [188]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
- Not matched
In the paper
“These efforts resemble the Partnership on AI's attempt to establish norms on model documentation processes [153].”
§ The Fallacy of AI Functionality > 5.2.2 Product Certification & Standards.
What the sources say
- Not matchedcites [153]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
- Not matched
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Collective industry-wide decision-making on critical issues can raise the bar for the entire industry and raise awareness within the industry of the importance of handling functionality challenges.”
§ The Fallacy of AI Functionality > 5.2.2 Product Certification & Standards.
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: fails — It may be too vague to check against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence. The paper’s conclusions may not depend on it. The paper may give no reason why it holds. It may state two separate claims as one.
In the paper
“Based on the FTC's approach to data security, in which the Commission has sued companies for failing to adequately secure consumer data in their possession against unknown thirdparty attackers [121], even post-deployment failures-if foreseeable and harmful-can be included among unfair practices, though they partially attributable to external actors.”
§ The Fallacy of AI Functionality > 5.1.1 Consumer Protection.
What the sources say
- Not matchedcites [121]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plainconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Not matched
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“The plaintiff need not have purchased or used the product; it is enough that they were injured by it, and the product has a defect that rendered it unsafe.”
§ The Fallacy of AI Functionality > 5.1.2 Products Liability Law.
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Due to clear conflicts of interest, the self-regulatory approaches described are far from adequate oversight for these challenges, and the presence of regulation does a lot to incentivise organizations to take these actions in the first place.”
§ The Fallacy of AI Functionality > 5.2 Organizational interventions
boosteddirect- Asserted: passes — The authors put it forward as true.
- Checkable: fails — It may be too vague to check against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence. The paper may give no reason why it holds. It may state two separate claims as one.
In the paper
“Similarly, the transportation industry has a rich history of thorough accident reports and investigations, through organizations such as the National Transportation and Safety Board (NTSB), who have already been responsible for assessing the damage from the few known cases of self-driving car crashes from Uber and Tesla [81].”
§ The Fallacy of AI Functionality > 5 DEALING WITH DYSFUNCTION: OPPORTUNITIES FOR INTERVENTION ON FUNCTIONAL SAFETY
What the sources say
- Not matchedcites [81]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Not matched
None of the sources cited for this claim could be matched to a paper in Crossref, so there is nothing to check it against.
In the paper
“Deceptive claims can be expressed explicitly-for example, representation in the sales materials that is inaccurate-or implied, such as an aspect of the design that suggests a functionality the product lacks [82,93].”
§ The Fallacy of AI Functionality > 5.1.1 Consumer Protection.
What the sources say
- Not matchedcites [82]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites [93]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
- Not matched
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“However, they do provide an immediate path forward in addressing these issues.”
§ The Fallacy of AI Functionality > 5.2 Organizational interventions
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds.
None of the sources cited for this claim could be matched to a paper in Crossref, so there is nothing to check it against.
In the paper
“In broader terms, the concept of functional safety in engineering design literature [163,174] well encapsulates the concerns articulated in this paper-namely that a system can be deployed without working very well, and that such performance issues can cause harm worth preventing.”
§ The Fallacy of AI Functionality > 5 DEALING WITH DYSFUNCTION: OPPORTUNITIES FOR INTERVENTION ON FUNCTIONAL SAFETY
What the sources say
- Not matchedcites [163]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites [174]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
boosteddirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may state two separate claims as one.
- Not matched
In the paper
“For software of nontrivial complexity, it is provably impossible to guarantee bug-free code.”
§ The Fallacy of AI Functionality > 5.1.2 Products Liability Law.
What the sources say
- Not matchedcites [36]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
boosted- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim.
- Not matched
In the paper
“Writing about software and tort law, Choi [36] identifies the complexity of software as a more fundamental type of hurdle.”
§ The Fallacy of AI Functionality > 5.1.2 Products Liability Law.
What the sources say
- Not matchedcites [36]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
- Not matched
In the paper
“One particular state law worth noting is California's Unfair Competition Law, which allows individuals to sue for injunctive relief to halt conduct that violates other laws, even if individuals could not otherwise sue under that law [2].”
§ The Fallacy of AI Functionality > 5.1.1 Consumer Protection.
What the sources say
- Not matchedcites [2]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
- Not matched
In the paper
“The Consumer Finance Protection Bureau can regulate harms from products dealing with loans, banking, or other consumer finance issues [4].”
§ The Fallacy of AI Functionality > 5.1.1 Consumer Protection.
What the sources say
- Not matchedcites [4]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
- Not matched
None of the sources cited for this claim could be matched to a paper in Crossref, so there is nothing to check it against.
In the paper
“Formally, the FTC does not have the power to issue fines under its Section 5 authority, but the Commission frequently enters into long-term consent decrees with companies that it sues, permitting continuing jurisdiction, monitoring, and fines for future violations [3,40].”
§ The Fallacy of AI Functionality > 5.1.1 Consumer Protection.
What the sources say
- Not matchedcites [3]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites [40]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Not matched
None of the sources cited for this claim could be matched to a paper in Crossref, so there is nothing to check it against.
In the paper
“On the U.S. front, industry groups IEEE and ISO regularly shape conversations, with IEEE going so far as to attempt the development of a certification program [72,84].”
§ The Fallacy of AI Functionality > 5.2.2 Product Certification & Standards.
What the sources say
- Not matchedcites [72]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites [84]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
- Not matched
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“The Federal Trade Commission is the federal consumer protection agency within the United States with the broadest subject matter jurisdiction.”
§ The Fallacy of AI Functionality > 5.1.1 Consumer Protection.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The challenge with such audits, however, is that the results are rarely communicated externally and disclosure is not mandatory, nor is it incentivized.”
§ The Fallacy of AI Functionality > 5.2.1 Internal Audits & Documentation.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: unclear — It is unclear why the paper says it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“A lack of construct validity would easily subject a model's user to liability.”
§ The Fallacy of AI Functionality > 5.1.5 Other Legal Avenues Already Being Explored.
hedgeddirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds.
None of the sources cited for this claim could be matched to a paper in Crossref, so there is nothing to check it against.
In the paper
“But as no amount of testing can guarantee bug-free software, it will difficult to determine how much testing is enough to be considered reasonable or non-negligent [36,94].”
§ The Fallacy of AI Functionality > 5.1.2 Products Liability Law.
What the sources say
- Not matchedcites [36]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites [94]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may state two separate claims as one.
- Not matched
The paper relies on this without support.
The iterative tagging procedure reliably and validly categorizes each case into the failure modes of the taxonomy.
“We split up the rows in the resulting set and used an iterative tagging procedure to come up with categories that associate each example with a different element or cause of failure.We updated, merged, and grouped our tags in meetings between tagging sessions, resulting in the following taxonomy.”
The paper relies on this without support.
The selected case studies from media and academic literature accurately illustrate and characterize the failure modes they are assigned to.
“We then chose known case studies from the media and academic literature to illustrate and best characterize these failure modes.”
The paper relies on this without support.
Poorly vetted AI products permeate our lives.
“Poorly vetted products permeate our lives, and while many readily accept the potential for harms as a tradeoff, the claims of the products' benefits go unchallenged.”
The paper relies on this without support.
The claims of AI products' benefits go unchallenged.
“Poorly vetted products permeate our lives, and while many readily accept the potential for harms as a tradeoff, the claims of the products' benefits go unchallenged.”
The paper relies on this without support.
AI operates as a general purpose technology prevalent in many industries.
“Thankfully, as AI operates as a general purpose technology prevalent in many of these industries, there already exists a plethora of governance infrastructure to address this issue in related fields of application.”
The paper relies on this without support.
Functionality is a safety risk aligned with profit-oriented goals.
“Also, as functionality is a safety risk aligned with profit-oriented goals, many model logging and evaluation operations tools are available for organizations to leverage in the internal inspection of their systems -including tools for more continuous monitoring of deployed systems [154,169].”
The paper relies on this without support.
AI hype will continue absent effective regulation.
“Communication failures often co-occur with other functional safety problems, and the lack of accountability for false claims -intentional or otherwise -makes these particularly pernicious and likely to occur as AI hype continues absent effective regulation.”
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“past work has highlighted the downstream harms made possible by model hallucinations, including in AI-powered search tools (Raji et al., 2022), and in code generation”
From HALoGEN: Fantastic LLM Hallucinations and Where to Find Them (2025) · cites this paper as Raji et al., 2022
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Algorithmic employment tools are notoriously unreliable and ill-defined, imposing significant harm on workers [46, 53, 54].”
From Null Compliance: NYC Local Law 144 and the challenges of algorithm accountability (2024) · cites this paper as [54]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“While there are multiple studies on the use of tools for fairness evaluation [54, 70, 119] and explainability [1, 46, 72, 82, 85, 98], fewer studies examine how practitioners evaluate other criteria such as basic functionality [117], safety, privacy, or recourse, just as fewer tools exist for this purpose.”
From Towards AI Accountability Infrastructure: Gaps and Opportunities in AI Audit Tooling (2025) · cites this paper as [117]
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“or lead to adverse effects from deploying suboptimal or harmful models (Bender and Friedman, 2018) on tasks for which they are ill-suited (Raji et al., 2022).”
From Lessons from the Trenches on Reproducible Evaluation of Language Models (2024) · cites this paper as Raji et al., 2022
The cited source says what the claim attributes to it.
In the citing paper
“The tools we surveyed in this category often overlooked entire other areas of concern, including the basic functionality of the model [117] as well as other methods of evaluation.”
From Towards AI Accountability Infrastructure: Gaps and Opportunities in AI Audit Tooling (2025) · cites this paper as [117]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ The Fallacy of AI Functionality > 1 INTRODUCTION
“AI, Algorithmic and Automation Incident and Controversy Repository (AAAIRC)”
§ The Fallacy of AI Functionality > 4.1 Methodology
“AI, Algorithmic and Automation Incident and Controversy Repository (AIAAIC)”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ The Fallacy of AI Functionality > Model Implementation Failures.
“the Epic sepsis prediction model, a product actually implemented at hundreds of hospitals, was recently externally validated by Wong et al. [203]”
§ The Fallacy of AI Functionality > References
“203. Laure Wynants, Ben Van Calster, Gary S Collins, Richard D Riley, Georg Heinze, Ewoud Schuit, Marc MJ Bonten, Darren L Dahly, Johanna A Damen, Thomas PA Debray, et al. 2020. Prediction models for diagnosis and prognosis of covid-19: systematic review and critical appraisal. bmj 369 (2020).”
Citations
Cites
Explaining and Harnessing Adversarial Examples
Goodfellow, I. J., Shlens, J., & Szegedy, C. (2014) · arXiv · doi:10.48550/arxiv.1412.6572 · Cited by 10,381
Dissecting racial bias in an algorithm used to manage the health of populations
Obermeyer, Z., Powers, B., Vogeli, C., & Mullainathan, S. (2019) · Science · doi:10.1126/science.aax2342 · Cited by 7,137
On the Dangers of Stochastic Parrots: Can Language Models Be Too Big? 🦜
Bender, E. M., Gebru, T., McMillan-Major, A., & Shmitchell, S. (2021) · Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency · doi:10.1145/3442188.3445922 · Cited by 6,988
The global landscape of AI ethics guidelines
Jobin, A., Ienca, M., & Vayena, E. (2019) · Nature Machine Intelligence · doi:10.1038/s42256-019-0088-2 · Cited by 6,220
(2018) · Cited by 3,510
Automating Inequality: How High-Tech Tools Profile, Police, and Punish the Poor
Lebovits, H. (2018) · Public Integrity · doi:10.1080/10999922.2018.1511671 · Cited by 3,320
Wynants, L., Van Calster, B., Bonten, M. M., Collins, G. S., Debray, T. P., De Vos, M., Haller, M. C., Heinze, G., Moons, K. G., Riley, R. D., et al. (2020) · doi:10.1101/2020.03.24.20041020 · Cited by 3,070
Barocas, S., & Selbst, A. D. (2014) · doi:10.2139/ssrn.2477899 · Cited by 3,048
Self-citationRaji, I. D., Smart, A., White, R. N., Mitchell, M., Gebru, T., Hutchinson, B., Smith-Loud, J., Theron, D., & Barnes, P. (2020) · Proceedings of the 2020 Conference on Fairness, Accountability, and Transparency · doi:10.1145/3351095.3372873 · Cited by 2,245
Self-citationCertifying and Removing Disparate Impact
Feldman, M., Friedler, S. A., Moeller, J., Scheidegger, C., & Venkatasubramanian, S. (2015) · Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining · doi:10.1145/2783258.2783311 · Cited by 1,827
Research works printed but not linked: 31
- 2013. Zhang v. Superior Ct., 304 P.3d 163 (2013).
- 2019. Stipulated Order for Civil Penalty, Monetary Judgment, and Injunctive Relief, No. 1:19-cv-2184, Docket 2-1 (D.D.C. July 24, 2019) (fining Facebook $5 billion for violating a prior consent decree).
- ACLU. 2018. ACLU Comment on New Amazon Statement Responding to Face Recognition Technology Test. https://www.aclu.org/press-releases/aclu-comment-new-amazon-statement-responding-face-recognition-technology-test. Accessed: 2022-1-12.
- Paul Berger. 2019. MTA’s Initial Foray Into Facial Recognition at High Speed Is a Bust. The Wall Street Journal(2019).
- Joseph Bernstein. 2021. Bad News. https://harpers.org/archive/2021/09/bad-news-selling-the-story-of-disinformation/. Harper’s Magazine (2021).
- Bryan H Choi. 2019. Crashworthy code. Wash. L. Rev. 94(2019), 39.
- Ángel Díaz and Laura Hecht. 2021. Double Standards in Social Media Content Moderation. https://www. brennancenter.org/sites/default/files/2021-08/Double_Standards_Content_Moderation.pdf. New York: Brennan Center for Justice(2021).
- Will Douglas Heaven. 2020. AI is wrestling with a replication crisis. MIT Technology Review (Nov. 2020).
- Alex C Engler. 2021. Independent auditors are struggling to hold AI companies accountable. FastCompany.
- 10.1145/2783258.2783311 · DOI · Paper page
- Clare Garvie. 2019. Garbage in, Garbage out. Face recognition on flawed data. Georgetown Law Center on Privacy & Technology (2019).
- M Harris. 2019. NTSB investigation into deadly Uber self-driving car crash reveals lax attitude toward safety. IEEE Spectrum (2019).
- Nicolaus Henke, Jordan Levine, and Paul McInerney. 2018. You Don’t Have to Be a Data Scientist to Fill This Must-Have Analytics Role. Harvard Business Review (Feb. 2018).
- Alex Hern. 2018. Cambridge Analytica: how did it turn clicks into votes. The Guardian 6(2018).
- Matthew Herper. 2017. MD Anderson Benches IBM Watson In Setback For Artificial Intelligence In Medicine. Forbes Magazine (Feb. 2017).
- F Patrick Hubbard. 2014. Sophisticated robots: balancing liability, regulation, and innovation. Fla. L. Rev. 66(2014), 1803.
- David Jeans. 2020. ScaleFactor Raised $100 Million In A Year Then Blamed Covid-19 For Its Demise. Employees Say It Had Much Bigger Problems. Forbes Magazine (July 2020).
- Frederike Kaltheuner, Abeba Birhane, Inioluwa Deborah Raji, Razvan Amironesei, Emily Denton, Alex Hanna, Hilary Nicole, Andrew Smart, Serena Dokuaa Oduro, James Vincent, Alexander Reben, Gemma Milne, Crofton Black, Adam Harvey, Andrew Strait, Tulsi Parida, Aparna Ashok, Fieke Jansen, Corinne Cath, and Aidan Peppin. 2021. Fake AI. Meatspace Press.
- Margot E Kaminski and Jennifer M Urban. 2021. The right to contest AI. Columbia Law Review 121, 7 (2021), 1957–2048.
- Lauren Kirchner and Matthew Goldstein. 2020. Access Denied: Faulty Automated Background Checks Freeze Out Renters. The Markup (2020).
- Lauren Kirchner and Matthew Goldstein. 2020. How Automated Background Checks Freeze Out Renters. The New York Times 28 (May 2020).
- Kumba Kpakima. 2021. Tiktok’s algorithm reportedly bans creators using terms ’Black’ and ’BLM’. https://i-d.vice.com/en_uk/article/m7epya/tiktoks-algorithm-reportedly-bans-creators-using-terms-black-and-blm. The Verge (2021).
- Colin Lecher. 2018. What happens when an algorithm cuts your health care. The Verge (2018).
- David Lehr and Paul Ohm. [n.d.]. Playing with the data: What legal scholars should learn about machine learning. https://lawreview.law.ucdavis.edu/issues/51/2/Symposium/51-2_Lehr_Ohm.pdf. Accessed: 2021-8-10.
- National Audit Office. 2020. Investigation into the response to cheating in English language tests - national audit office (NAO) press release. https://www.nao.org.uk/press-release/investigation-into-the-response-to-cheating-in-english-language-tests/
- Jesse O’Neill. 2021. Facebook cracks down on discussing ‘hoes’ in gardening group. https://nypost.com/2021/07/20/facebook-cracks-down-on-discussing-hoes-in-gardening-group/.
- Rashida Richardson, Jason Schultz, and Kate Crawford. 2019. Dirty Data, Bad Predictions: How Civil Rights Violations Impact Police Data, Predictive Policing Systems, and Justice. (Feb. 2019).
- Rashida Richardson, Jason M Schultz, and Vincent M Southerland. 2019. Litigating Algorithms: 2019 US Report. AI Now Institute, September(2019).
- David S Rubenstein. 2021. Acquiring ethical AI. Florida Law Review 73(2021).
- J Shane. 2019. Janelle Shane: The danger of AI is weirder than you think TED Talk, 10: 20. Katsottu 8.8, 2020.
- Luke Stark and Jevan Hutson. 2022. Physiognomic Artificial Intelligence. forthcoming in Fordham Intellectual Property, Media & Entertainment Law Journal XXXII (2022). https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3927300
Cited by
Sorted by citations
Sociotechnical Harms of Algorithmic Systems: Scoping a Taxonomy for Harm Reduction
Shelby, R., Rismani, S., Henne, K., Moon, A., Rostamzadeh, N., Nicholas, P., Yilla-Akbari, N., Gallegos, J., Smart, A., Garcia, E., et al. (2023) · Proceedings of the 2023 AAAI/ACM Conference on AI, Ethics, and Society · doi:10.1145/3600211.3604673 · Cited by 279
Auditing large language models: a three-layered approach
Mökander, J., Schuett, J., Kirk, H. R., & Floridi, L. (2024) · AI and Ethics · doi:10.1007/s43681-023-00289-2 · Cited by 289
Data science, artificial intelligence and the third wave of digital era governance
Dunleavy, P., & Margetts, H. (2023) · doi:10.31219/osf.io/f3rza · Cited by 186
Hallucination‐Free? Assessing the Reliability of Leading AI Legal Research Tools
Magesh, V., Surani, F., Dahl, M., Suzgun, M., Manning, C. D., & Ho, D. E. (2025) · Journal of Empirical Legal Studies · doi:10.1111/jels.12413 · Cited by 182
When Human-AI Interactions Become Parasocial: Agency and Anthropomorphism in Affective Design
Maeda, T., & Quan-Haase, A. (2024) · The 2024 ACM Conference on Fairness, Accountability, and Transparency · doi:10.1145/3630106.3658956 · Cited by 144
Metadata
- Authors
- Raji, I. D., Kumar, I. E., Horowitz, A., & Selbst, A.
- Affiliations
- University of California, Berkeley (US)
- Brown University (US)
- American Civil Liberties Union (US)
- University of California, Los Angeles (US)
- Venue
- 2022 ACM Conference on Fairness Accountability and Transparency
- Publisher
- ACM
- Published
- 2022-06-20
- Pages
- 959-972
- OpenAlex
- W4283156811
- Cited by
- 251
- References
- 130
- Citations by year
- 2026: 50 · 2025: 93 · 2024: 64 · 2023: 40 · 2022: 4
- FWCI
- 49.94
- Type
- conference paper
- Language
- en
- Open access
- bronze
- doi.org
- Locations
- ACM Conference on Fairness, Accountability, and Transparency (FAccT) (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University) (cc-by)
- Funders
- Mozilla Foundation
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Ethics and Social Impacts of AI
- Social Sciences › Law › Digitalization, Law, and Regulation
- Physical Sciences › Computer Science › Artificial Intelligence › Explainable Artificial Intelligence (XAI)
- Keywords
- AI policy, algorithmic harms, value alignment
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
A language model drew the claims from the paper on 2026-10-30, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-30 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
