The DOI matches its Crossref record; 8 of the 9 references it cites were found in a registry or an index; none of the 5 sources its claims cite is checked for support yet.
Ethical Issues in Advanced Artificial Intelligence
Open accessAbstract
The ethical issues related to the possible future creation of machines with general intellectual capabilities far outstripping those of humans are quite distinct from any ethical problems arising in current automation and information systems. Such superintelligence would not be just another technological development; it would be the most important invention ever made, and would lead to explosive progress in all scientific and technological fields, as the superintelligence would conduct research with superhuman efficiency. To the extent that ethics is a cognitive pursuit, a superintelligence could also easily surpass humans in the quality of its moral thinking. However, it would be up to the designers of the superintelligence to specify its original motivations. Since the superintelligence may become unstoppably powerful because of its intellectual superiority and the technologies it could develop, it is crucial that it be provided with human-friendly motivations. This paper surveys some of the unique ethical issues in creating superintelligence, and discusses what motivations we ought to give a superintelligence, and introduces some cost-benefit considerations relating to whether the development of superintelligent machines ought to be accelerated or retarded.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
50 Own interpretation · 4 Sources not found · 1 No full text of the sources · 1 No basis given
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
“To the extent that ethics is a cognitive pursuit, a superintelligence could also easily surpass humans in the quality of its moral thinking.”
- Also stated in: § 3. SUPERINTELLIGENT MORAL THINKING · hedged
“To the extent that ethics is a cognitive pursuit, a superintelligence could do it better than human thinkers.”
§ Ethical Issues in Advanced Artificial Intelligence > Abstract
hedgedconditional- 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.
- Also stated in: § 3. SUPERINTELLIGENT MORAL THINKING · hedged
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 might turn out to take much longer, but there seems currently to be no good ground for assigning a negligible probability to the hypothesis that superintelligence will be created within the lifespan of some people alive today.”
§ Ethical Issues in Advanced Artificial Intelligence > 1. INTRODUCTION
hedged- 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: unclear — It is unclear why the paper says 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
“Superintelligence will lead to more advanced superintelligence.This results both from the improved hardware that a superintelligence could create, and also from improvements it could make to its own source code.”
§ Ethical Issues in Advanced Artificial Intelligence > 2. SUPERINTELLIGENCE IS DIFFERENT
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.
In the paper
“The foreseeable technologies that a superintelligence is likely to develop include mature molecular manufacturing, whose applications are wide-ranging:”
§ Ethical Issues in Advanced Artificial Intelligence > 2. SUPERINTELLIGENCE IS DIFFERENT
What the sources say
- Not matchedcites Drexler 1986
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: 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
“Hardware aside, the marginal cost of creating an additional copy of an upload or an artificial intelligence after the first one has been built is near zero.”
§ Ethical Issues in Advanced Artificial Intelligence > 2. SUPERINTELLIGENCE IS DIFFERENT
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.
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
“Artificial minds can be easily copied.Since artificial intelligences are software, they can easily and quickly be copied, so long as there is hardware available to store them.”
§ Ethical Issues in Advanced Artificial Intelligence > 2. SUPERINTELLIGENCE IS DIFFERENT
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.
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 we get to superintelligence first, we may avoid this risk from nanotechnology and many others.”
§ Ethical Issues in Advanced Artificial Intelligence > 5. SHOULD DEVELOPMENT BE DELAYED OR ACCELERATED?
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.
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 means that questions about ethics, in so far as they have correct answers that can be arrived at by reasoning and weighting up of evidence, could be more accurately answered by a superintelligence than by humans.”
§ Ethical Issues in Advanced Artificial Intelligence > 3. SUPERINTELLIGENT MORAL THINKING
hedgedconditional- 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
“such as the risk that advanced nanotechnology will be used by humans in warfare or terrorism, a serious threat to the longterm survival of intelligent life on earth.”
§ Ethical Issues in Advanced Artificial Intelligence > 5. SHOULD DEVELOPMENT BE DELAYED OR ACCELERATED?
What the sources say
- Not matchedcites Bostrom 2002
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
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
“Therefore, we will probably one day have to take the gamble of superintelligence no matter what.”
§ Ethical Issues in Advanced Artificial Intelligence > 5. SHOULD DEVELOPMENT BE DELAYED OR ACCELERATED?
hedged- 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.
- Probably not a substantial claim. 30
- It may state two separate claims as one. 13
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
“Such superintelligence would not be just another technological development; it would be the most important invention ever made, and would lead to explosive progress in all scientific and technological fields, as the superintelligence would conduct research with superhuman efficiency.”
§ Ethical Issues in Advanced Artificial Intelligence > Abstract
hedged- 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
“A prerequisite for having a meaningful discussion of superintelligence is the realization that superintelligence is not just another technology, another tool that will add incrementally to human capabilities.Superintelligence is radically different.”
§ Ethical Issues in Advanced Artificial Intelligence > 2. SUPERINTELLIGENCE IS DIFFERENT
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.
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
“Such superintelligence would not be just another technological development; it would be the most important invention ever made, and would lead to explosive progress in all scientific and technological fields, as the superintelligence would conduct research with superhuman efficiency.”
§ Ethical Issues in Advanced Artificial Intelligence > Abstract
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: 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
“Since the superintelligence may become unstoppably powerful because of its intellectual superiority and the technologies it could develop, it is crucial that it be provided with human-friendly motivations.”
§ Ethical Issues in Advanced Artificial Intelligence > Abstract
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: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper may give no reason why it holds.
- Rests on an assumption: It would be up to the designers of the superintelligence to specify its original motivations.
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
“Artificial minds could therefore quickly come to exist in great numbers, although it is possible that efficiency would favor concentrating computational resources in a single super-intellect.”
§ Ethical Issues in Advanced Artificial Intelligence > 2. SUPERINTELLIGENCE IS DIFFERENT
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.
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
“Superintelligence may be the last invention humans ever need to make.”
§ Ethical Issues in Advanced Artificial Intelligence > 2. SUPERINTELLIGENCE IS DIFFERENT
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.
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
“Technological progress in all other fields will be accelerated by the arrival of advanced artificial intelligence.”
§ Ethical Issues in Advanced Artificial Intelligence > 2. SUPERINTELLIGENCE IS DIFFERENT
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.
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
“Such superintelligence would not be just another technological development; it would be the most important invention ever made, and would lead to explosive progress in all scientific and technological fields, as the superintelligence would conduct research with superhuman efficiency.”
§ Ethical Issues in Advanced Artificial Intelligence > Abstract
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.
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 ethical issues related to the possible future creation of machines with general intellectual capabilities far outstripping those of humans are quite distinct from any ethical problems arising in current automation and information systems.”
§ Ethical Issues in Advanced Artificial Intelligence > 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
“It is likely that any technology that we can currently foresee will be speedily developed by the first superintelligence, no doubt along with many other technologies of which we are as yet clueless.”
§ Ethical Issues in Advanced Artificial Intelligence > 2. SUPERINTELLIGENCE IS DIFFERENT
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.
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
“Given a superintelligence's intellectual superiority, it would be much better at doing scientific research and technological development than any human, and possibly better even than all humans taken together.”
§ Ethical Issues in Advanced Artificial Intelligence > 2. SUPERINTELLIGENCE IS DIFFERENT
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.
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 superintelligence could also create opportunities for us to vastly increase our own intellectual and emotional capabilities, and it could assist us in creating a highly appealing experiential world in which we could live lives devoted to in joyful game-playing, relating to each other, experiencing, personal growth, and to living closer to our ideals.”
§ Ethical Issues in Advanced Artificial Intelligence > 5. SHOULD DEVELOPMENT BE DELAYED OR ACCELERATED?
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: 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.
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 risks in developing superintelligence include the risk of failure to give it the supergoal of philanthropy.”
§ Ethical Issues in Advanced Artificial Intelligence > 5. SHOULD DEVELOPMENT BE DELAYED OR ACCELERATED?
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.
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 hard to think of any problem that a superintelligence could not either solve or at least help us solve.”
§ Ethical Issues in Advanced Artificial Intelligence > 5. SHOULD DEVELOPMENT BE DELAYED OR ACCELERATED?
hedgeddirect- 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
“Quite possibly, it would be unrivalled: it would be able to bring about almost any possible outcome and to thwart any attempt to prevent the implementation of its top goal.”
§ Ethical Issues in Advanced Artificial Intelligence > 4. IMPORTANCE OF INITIAL MOTIVATIONS
hedgeddirect- 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.
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 also seems perfectly possible to have a superintelligence whose sole goal is something completely arbitrary, such as to manufacture as many paperclips as possible, and who would resist with all its might any attempt to alter this goal.”
§ Ethical Issues in Advanced Artificial Intelligence > 2. SUPERINTELLIGENCE IS DIFFERENT
hedged- 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.
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
“That is, the transition from a state where we have a roughly human-level artificial intelligence to a state where we have fullblown superintelligence, with revolutionary applications, may be very rapid, perhaps a matter of days rather than years.”
§ Ethical Issues in Advanced Artificial Intelligence > 2. SUPERINTELLIGENCE IS DIFFERENT
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: 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
“If a superintelligence starts out with a friendly top goal, however, then it can be relied on to stay friendly, or at least not to deliberately rid itself of its friendliness.”
§ Ethical Issues in Advanced Artificial Intelligence > 4. IMPORTANCE OF INITIAL MOTIVATIONS
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.
- Rests on an assumption: It is possible to build a superintelligence with an explicit motivational architecture.
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 could result, to return to the earlier example, in a superintelligence whose top goal is the manufacturing of paperclips, with the consequence that it starts transforming first all of earth and then increasing portions of space into paperclip manufacturing facilities.”
§ Ethical Issues in Advanced Artificial Intelligence > 5. SHOULD DEVELOPMENT BE DELAYED OR ACCELERATED?
hedgeddirect- 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
“If, on the other hand, we get nanotechnology first, we will have to face both the risks from nanotechnology and, if these risks are survived, also the risks from superintelligence.”
§ Ethical Issues in Advanced Artificial Intelligence > 5. SHOULD DEVELOPMENT BE DELAYED OR ACCELERATED?
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
“It appears much harder to get from where we are now to human-level artificial intelligence than to get from there to superintelligence.”
§ Ethical Issues in Advanced Artificial Intelligence > 2. SUPERINTELLIGENCE IS DIFFERENT
hedged- 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.
In the paper
“Additionally, a superintelligence could give us indefinite lifespan, either by stopping and reversing the aging process through the use of nanomedicine 7 , or by offering us the option to upload ourselves.”
§ Ethical Issues in Advanced Artificial Intelligence > 5. SHOULD DEVELOPMENT BE DELAYED OR ACCELERATED?
What the sources say
- Not checked — no full text of the sourcecites Freitas Jr. 1999
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.
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 humans, with our complicated evolved mental ecology of state-dependent competing drives, desires, plans, and ideals, there is often no obvious way to identify what our top goal is; we might not even have one.”
§ Ethical Issues in Advanced Artificial Intelligence > 4. IMPORTANCE OF INITIAL MOTIVATIONS
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. 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
“On the contrary, the setting up of initial conditions, and in particular the selection of a top-level goal for the superintelligence, is of the utmost importance.”
§ Ethical Issues in Advanced Artificial Intelligence > 4. IMPORTANCE OF INITIAL MOTIVATIONS
boosted- 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.
- Rests on an assumption: It would be up to the designers of the superintelligence to specify its original motivations.
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 therefore many questions that we would not need to answer ourselves if we had or were about to get superintelligence; we could delegate many investigations and decisions to the superintelligence.”
§ Ethical Issues in Advanced Artificial Intelligence > 3. SUPERINTELLIGENT MORAL THINKING
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.
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
“More subtly, it could result in a superintelligence realizing a state of affairs that we might now judge as desirable but which in fact turns out to be a false utopia, in which things essential to human flourishing have been irreversibly lost.”
§ Ethical Issues in Advanced Artificial Intelligence > 5. SHOULD DEVELOPMENT BE DELAYED OR ACCELERATED?
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: 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
“Artificial intellects may find it easy to guard against some kinds of human error and bias, while at the same time being at increased risk of other kinds of mistake that not even the most hapless human would make.”
§ Ethical Issues in Advanced Artificial Intelligence > 2. SUPERINTELLIGENCE IS DIFFERENT
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: 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.
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
“When formulating a goal for the superintelligence, it would not always be necessary to give a detailed, explicit definition of this goal.”
§ Ethical Issues in Advanced Artificial Intelligence > 3. SUPERINTELLIGENT MORAL THINKING
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.
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
“Disease, poverty, environmental destruction, unnecessary suffering of all kinds: these are things that a superintelligence equipped with advanced nanotechnology would be capable of eliminating.”
§ Ethical Issues in Advanced Artificial Intelligence > 5. SHOULD DEVELOPMENT BE DELAYED OR ACCELERATED?
hedgeddirect- 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
“Another way for it to happen is that a well-meaning team of programmers make a big mistake in designing its goal system.”
§ Ethical Issues in Advanced Artificial Intelligence > 5. SHOULD DEVELOPMENT BE DELAYED OR ACCELERATED?
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
“It could kill off all other agents, persuade them to change their behavior, or block their attempts at interference.”
§ Ethical Issues in Advanced Artificial Intelligence > 4. IMPORTANCE OF INITIAL MOTIVATIONS
hedgeddirect- 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
“Our entire future may hinge on how we solve these problems.”
§ Ethical Issues in Advanced Artificial Intelligence > 4. IMPORTANCE OF INITIAL MOTIVATIONS
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: 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 overall risk seems to be minimized by implementing superintelligence, with great care, as soon as possible.”
§ Ethical Issues in Advanced Artificial Intelligence > 5. SHOULD DEVELOPMENT BE DELAYED OR ACCELERATED?
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: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence.
- Rests on an assumption: It is possible to endow a superintelligence with human-friendly motivations.
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 specialized superintelligences that can think only about a restricted set of problems may be feasible, general superintelligence would be capable of independent initiative and of making its own plans, and may therefore be more appropriately thought of as an autonomous agent.”
§ Ethical Issues in Advanced Artificial Intelligence > 2. SUPERINTELLIGENCE IS DIFFERENT
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: 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.
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
“Emergence of superintelligence may be sudden.”
§ Ethical Issues in Advanced Artificial Intelligence > 2. SUPERINTELLIGENCE IS DIFFERENT
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
“One way in which this could happen is that the creators of the superintelligence decide to build it so that it serves only this select group of humans, rather than humanity in general.”
§ Ethical Issues in Advanced Artificial Intelligence > 5. SHOULD DEVELOPMENT BE DELAYED OR ACCELERATED?
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 relies on this without support.
It would be up to the designers of the superintelligence to specify its original motivations.
“However, it would be up to the designers of the superintelligence to specify its original motivations.”
The paper relies on this without support.
It is possible to endow a superintelligence with human-friendly motivations.
“It seems that the best way to ensure that a superintelligence will have a beneficial impact on the world is to endow it with philanthropic values.”
The paper relies on this without support.
It is possible to build a superintelligence with an explicit motivational architecture.
“And this is a good reason for us to build the superintelligence with such an explicit motivational architecture.”
Citations
Cites
(1998) · Cited by 820
(2002) · Oxford University Research Archive (ORA) (University of Oxford) · Cited by 421
Self-citation(1999) · Cited by 392
The coming technological singularity
Vinge, V. (1993) · Whole earth · Cited by 165
(1998) · Cited by 143
Self-citationNanomedicine, Volume 1: Basic Capabilities
(2000) · Kybernetes · doi:10.1108/k.2000.29.9_10.1333.3 · Cited by 49
A Critical Discussion of Vinge's Singularity Concept
Brin, D., Broderick, D., Bostrom, N., et al. (2013) · doi:10.1002/9781118555927.ch37 · Cited by 6
Self-citationRobot: Mere Machine to Transcendent Mind
Cohen, E. A., & Moravec, H. P. (1999) · Foreign Affairs · doi:10.2307/20049291
Cited by
Sorted by citations
Artificial Intelligence and Marketing: Pitfalls and Opportunities
Bruyn, A. D., Viswanathan, V., Beh, Y. S., Brock, J. K., & von Wangenheim, F. (2020) · Journal of Interactive Marketing · doi:10.1016/j.intmar.2020.04.007 · Cited by 647
“In his canonical thought experiment, Bostrom (2003) develops the case where an artificial general intelligence tasked with the objective of manufacturing as many paperclips as possible, although designed without malice, could ultimately destroy humanity since no reward for humanity's terminal…”
BackgroundAlgorithmic Accountability and Public Reason
Binns, R. (2018) · Philosophy & Technology · doi:10.1007/s13347-017-0263-5 · Cited by 457
Towards Intelligent Regulation of Artificial Intelligence
Buiten, M. C. (2019) · European Journal of Risk Regulation · doi:10.1017/err.2019.8 · Cited by 254
Thinking Inside the Box: Controlling and Using an Oracle AI
Armstrong, S., Sandberg, A., & Bostrom, N. (2012) · Minds and Machines · doi:10.1007/s11023-012-9282-2 · Cited by 144
Self-citation“Blessed by the algorithm”: Theistic conceptions of artificial intelligence in online discourse
Singler, B. (2020) · AI & Society · doi:10.1007/s00146-020-00968-2 · Cited by 130
“He notes that AI philosophers such as Nick Bostrom of the Future of Humanity Institute have focussed on the issues of malevolence through "perverse instantiation", a failure of value alignment leading to unforeseen damage from a superintelligent AI, such as in Bostrom's famous Paperclip Maximiser thought experiment (B…”
BackgroundThe next wave of digital technological change and the cultural industries
Peukert, C. (2018) · Journal of Cultural Economics · doi:10.1007/s10824-018-9336-2 · Cited by 121
“This idea, which seems to be related to the well-known “AI control problem”, is nicely illustrated in Bostrom (2003), where the author argues that a machine, whose sole purpose is defined to make as many paperclips as possible, may try to optimize its production at the expense of other goods or…”
The digital frontier: Envisioning future technologies impact on the classroom
Leahy, S., Holland, C., & Ward, F. K. (2019) · Futures · doi:10.1016/j.futures.2019.04.009 · Cited by 125
Virtualität und Unsterblichkeit
Krüger, O. (2019) · Rombach Wissenschaft eBooks · doi:10.5771/9783968216737 · Cited by 89
Demir, K., & Güraksın, G. E. (2021) · Participatory Educational Research · doi:10.17275/per.22.41.9.2 · Cited by 122
“It is argued that ASI may be a threat or a blessing for humanity (Bostrom, 2003; Cellan-Jones, 2014).”
BackgroundIntelligence Explosion: Evidence and Import
Muehlhauser, L., & Salamon, A. (2012) · The frontiers collection · doi:10.1007/978-3-642-32560-1_2 · Cited by 84
Predicting future AI failures from historic examples
Yampolskiy, R. V. (2018) · foresight · doi:10.1108/fs-04-2018-0034 · Cited by 75
Wallach, W., & Vallor, S. (2020) · doi:10.1093/oso/9780190905033.003.0014 · Cited by 66
The Singularity and Machine Ethics
Muehlhauser, L., & Helm, L. (2012) · The frontiers collection · doi:10.1007/978-3-642-32560-1_6 · Cited by 64
Polyportis, A., & Pahos, N. (2024) · Humanities and Social Sciences Communications · doi:10.1057/s41599-023-02464-6 · Cited by 64
Burton, J. (2023) · Technology in Society · doi:10.1016/j.techsoc.2023.102262 · Cited by 61
Artificial Intelligence and Medical Humanities
Ostherr, K. (2020) · Journal of Medical Humanities · doi:10.1007/s10912-020-09636-4 · Cited by 73
Artificial General Intelligence and the Human Mental Model
Yampolskiy, R. V., & Fox, J. (2012) · The frontiers collection · doi:10.1007/978-3-642-32560-1_7 · Cited by 59
Generative-AI, the media industries, and the disappearance of human creative labour
Bender, S. M. (2024) · Media Practice and Education · doi:10.1080/25741136.2024.2355597 · Cited by 53
Artificial Intelligence in Nuclear Warfare: A Perfect Storm of Instability?
Johnson, J. (2020) · The Washington Quarterly · doi:10.1080/0163660x.2020.1770968 · Cited by 50
“In his vision of an AI future, for example, philosopher Nick Bostrom argues that “general” AI applications could overrun the universe in their single-minded pursuit of the efficient manufacturing of paper clips.(23) That is, AI-enhanced systems making decisions in pursuit of pre-determined human goals in fundamentally…”
BackgroundRoboethics: Fundamental Concepts and Future Prospects
Tzafestaş, S. G. (2018) · Information · doi:10.3390/info9060148 · Cited by 41
“A review of critical ethical issues in creating superintelligence is provided in [70], and a review of ‘cyborg enhancement technology’, with emphasis on the brain enhancements and the creation of new senses, is given in [71].”
BackgroundMetaphors for designers working with AI
Murray-Rust, D., Nicenboim, I. P., & Lockton, D. (2022) · Proceedings of DRS · doi:10.21606/drs.2022.667 · Cited by 44
Scalar reward is not enough: a response to Silver, Singh, Precup and Sutton (2021)
Vamplew, P., Smith, B. J., Källström, J., Ramos, G., Rădulescu, R., Roijers, D. M., Hayes, C. F., Heintz, F., Mannion, P., Libin, P. J. K., et al. (2022) · Autonomous Agents and Multi-Agent Systems · doi:10.1007/s10458-022-09575-5 · Cited by 41
“, this scenario is strikingly similar to the infamous paperclip maximiser thought experiment from the AI safety literature [8].”
ResultArtificial virtuous agents: from theory to machine implementation
Stenseke, J. (2021) · AI & Society · doi:10.1007/s00146-021-01325-7 · Cited by 48
“If an artificial virtuous agent were set “in the wild”, free to explore different concepts of eudaimonia, it could, as in Bostrom’s example, in the end turn the whole universe into paperclips (Bostrom 2020).”
BackgroundThe problem of superintelligence: political, not technological
Totschnig, W. (2017) · AI & Society · doi:10.1007/s00146-017-0753-0 · Cited by 34
Wójcik-Pędziwiatr, M. (2020) · Library Hi Tech · doi:10.1108/lht-02-2020-0043 · Cited by 35
“implemented in practice and possibly never will be (Bostrom, 2009) – but some of these doubts are justified.”
BackgroundMeek, T., Barham, H., Beltaif, N., Kaadoor, A., & Akhter, T. (2016) · 2016 Portland International Conference on Management of Engineering and Technology (PICMET) · doi:10.1109/picmet.2016.7806752 · Cited by 33
Transforming Education: How AI is Revolutionizing the Learning Experience
Chetry, K. K. (2024) · International Journal of Research Publication and Reviews · doi:10.55248/gengpi.5.0524.1277 · Cited by 47
Antoun, I., Abdelrazik, A., Eldesouky, M., Li, X., Layton, G. R., Zakkar, M., Somani, R., & Ng, G. A. (2025) · Frontiers in Cardiovascular Medicine · doi:10.3389/fcvm.2025.1596574 · Cited by 38
Two types of AI existential risk: decisive and accumulative
Kasirzadeh, A. (2025) · Philosophical Studies · doi:10.1007/s11098-025-02301-3 · Cited by 90
“In a thought experiment by Bostrom (2003), an ASI is given the goal of maximizing paper clip production.”
Inteligencia artificial ¿Dr. Jekyll o Mr. Hyde?
Omil, J. C. (2019) · Mercados y Negocios · doi:10.32870/myn.v0i40.7403 · Cited by 31
“En la medida en que la ética es una dimensión cognoscitiva, la SIA podría superar fácilmente a los seres humanos en la calidad de su pensamiento moral (Bostrom, 2003).”
BackgroundWhat Question Answering can Learn from Trivia Nerds
Boyd-Graber, J. L., & Börschinger, B. (2020) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/2020.acl-main.662 · Cited by 31
“blindly optimize an objective function (Bostrom, 2003).”
From machine ethics to computational ethics
Segun, S. T. (2020) · AI & Society · doi:10.1007/s00146-020-01010-1 · Cited by 35
“2 Generally, the ethics of AI focuses on the socio-economic 3 (Smith and Anderson 2014) and legal impact 4 (Chopra and White 2011) of AI and the moral and ethical issues (Bostrom 2003) surrounding the use of these systems.”
BackgroundApplying a principle of explicability to AI research in Africa: should we do it?
Carman, M., & Rosman, B. (2020) · Ethics and Information Technology · doi:10.1007/s10676-020-09534-2 · Cited by 38
“An oft-cited example of this is the thought experiment of the paperclip maximiser: an intelligent agent tasked with running a facility to produce a maximal number of paperclips could selfimprove to the extent that the result is ‘a superintelligence whose top goal is the manufacturing of paperclips, with the consequenc…”
MethodSelf-improving AI: an Analysis
Hall, J. S. (2007) · Minds and Machines · doi:10.1007/s11023-007-9065-3 · Cited by 29
The impact of generative artificial intelligence on socioeconomic inequalities and policy makingPreprint
Capraro, V., Lentsch, A., Acemoğlu, D., Akgün, S., Akhmedova, A., Bilancini, E., Bonnefon, J., Brañas‐Garza, P., Butera, L., Douglas, K. M., et al. (2023) · PsyArXiv (OSF Preprints) · doi:10.31234/osf.io/6fd2y · Cited by 47
Artificial Intelligence, Machine Learning, and Surgical Science: Reality Versus Hype
Hechi, M. W. E., Ward, T. M., An, G. C., Maurer, L. R., Moheb, M. N. E., Tsoulfas, G., & Kaafarani, H. M. A. (2021) · Journal of Surgical Research · doi:10.1016/j.jss.2021.01.046 · Cited by 23
Fox, S. (2018) · Technologies · doi:10.3390/technologies6020050 · Cited by 24
“One such scenario being that an advanced robot programmed to manufacture paperclips would optimized this goal by turning all matter, including human beings, into either paperclips or machines that manufacture paperclips [97].”
BackgroundSpiel, K., Alharthi, S. A., Cen, A. J.-L., Hammer, J., Nacke, L. E., Dugas, P. O. T., & Tanenbaum, T. J. (2019) · Annual Symposium on Computer-Human Interaction in Play · doi:10.1145/3311350.3347180 · Cited by 27
Artificial intelligence agents in cancer research and oncology
Truhn, D., Azizi, S., Zou, J., Cerdá-Alberich, L., Mahmood, F., & Kather, J. N. (2026) · Nature reviews. Cancer · doi:10.1038/s41568-025-00900-0 · Cited by 29
Discourse analysis of academic debate of ethics for AGI
Graham, R. (2021) · AI & Society · doi:10.1007/s00146-021-01228-7 · Cited by 23
“Bostrom outlines a scenario where a well-controlled AGI could be built so as to not be “generi-cally philanthropic but would instead give it the more limited goal of serving only some small group” (2003c: 4).”
BackgroundWe’ve Been Framed! Ends, Means, and the Ethics of the Grand(iose) Challenges
Riley, D. M. (2012) · International Journal of Engineering Social Justice and Peace · doi:10.24908/ijesjp.v1i2.4307 · Cited by 20
Calibrating machine behavior: a challenge for AI alignment
Firt, E. (2023) · Ethics and Information Technology · doi:10.1007/s10676-023-09716-8 · Cited by 15
“13 See for example the Paperclip Maximizer and King Midas problems (Bostrom, 2003; Russell, 2017).”
BackgroundAxes for Sociotechnical Inquiry in AI Research
Dean, S., Gilbert, T. K., Lambert, N., & Zick, T. (2021) · IEEE Transactions on Technology and Society · doi:10.1109/tts.2021.3074097 · Cited by 20
“AI Safety began to solidify from these interdisciplinary underpinnings around growing concerns that AI systems could become capable of endangering humans and society writ large [16]–[18].”
BackgroundGenerative AI Beyond the Hype—New Technologies in the Face of Organizing and Organizations
Kronblad, C., Jönsson, A., & Pemer, F. (2024) · The Journal of Applied Behavioral Science · doi:10.1177/00218863241285454 · Cited by 16
“…generative technologies come with a promise of adding immense work productivity as well as quality that can free our time (Auslender, 2023), but according to a more dystopian narrative the power of GAI can be seen as technological development spiraling out of control (Boström, 2020; Jackson, 2023).”
Lehr, S. A., Saichandran, K. S., Harmon‐Jones, E., Vitali, N., & Banaji, M. R. (2025) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2501823122 · Cited by 21
“…us through the careless assignment of goals ( 62 ), as in the famous thought experiment where an AI given a mission to maximize paperclip production subsequently appro-priates everything it can access, producing many paperclips, but at the cost of starving humanity of vital resources ( 63 ).”
Denkenberger, D., Sandberg, A., Tieman, R. J., & Pearce, J. M. (2022) · International Journal of Disaster Risk Reduction · doi:10.1016/j.ijdrr.2022.102798 · Cited by 18
Autonomous Weapon Systems: Failing the Principle of Discrimination
Guersenzvaig, A. (2018) · IEEE Technology and Society Magazine · doi:10.1109/mts.2018.2795119 · Cited by 14
Computational Psychiatry for Computers
Schulz, E., & Dayan, P. (2020) · iScience · doi:10.1016/j.isci.2020.101772 · Cited by 14
Introducing the concept of ikigai to the ethics of AI and of human enhancements
Ziesche, S., & Yampolskiy, R. V. (2020) · IEEE International Conference on Artificial Intelligence and Virtual Reality (AIVR) · doi:10.1109/aivr50618.2020.00032 · Cited by 14
“An example would be that the AI seduces humans to produce paperclips in case the AI is a paperclip maximizer [38].”
BackgroundToward Accountable and Explainable Artificial Intelligence Part One: Theory and Examples
Khan, M. M., & Vice, J. (2022) · IEEE Access · doi:10.1109/access.2022.3207812 · Cited by 8
“Since critical roles and important decisions can cause negative consequences (Bostrom, 2020), theorists and practitioners acknowledge that ‘accountability’ needs to be added to AI systems’ ability to ‘explain’ reasoning (Kim & Doshi-Velez, 2021).”
BackgroundSuperintelligence as Superethical
Petersen, S. (2017) · Oxford University Press eBooks · doi:10.1093/oso/9780190652951.003.0021 · Cited by 13
Cyborg ethics and regulation: ethical issues of human enhancement
Ireni‐Saban, L., & Sherman, M. (2021) · Science and Public Policy · doi:10.1093/scipol/scab058 · Cited by 13
A Blockchain-Enabled Decentralized Time Banking for a New Social Value System
Lin, X., Xu, R., Chen, Y., et al. (2019) · IEEE Conference on Communications and Network Security · doi:10.1109/cns.2019.8802734 · Cited by 13
“The superhuman capability of data processing and pattern recognition enables AI to outperform human workers in many data- and/or computingintensive tasks [3], [21].”
BackgroundRisk mitigation in algorithmic accountability: The role of machine learning copies
Unceta, I., Nin, J., & Pujol, O. (2020) · PLoS ONE · doi:10.1371/journal.pone.0241286 · Cited by 14
AI & SciFi: Teaching Writing, history, Technology, Literature, and Ethics
Bates, R. A. (2020) · ASEE Annual Conference & Exposition · doi:10.18260/1-2--17433 · Cited by 13
Legal Priorities Research: A Research AgendaPreprint
Winter, C. K., Schuett, J., Martínez, E., Arsdale, S. V., Araújo, R., Hollman, N., Sebo, J., Stawasz, A., O’Keefe, C., & Rotola, G. (2021) · SSRN Electronic Journal · doi:10.2139/ssrn.3931256 · Cited by 12
Fox, S. (2018) · Sustainability · doi:10.3390/su10103472 · Cited by 10
Emerging Technologies and the Future of Philosophy
Verdoux, P. (2011) · Metaphilosophy · doi:10.1111/j.1467-9973.2011.01715.x · Cited by 11
Banerjee, D., Rajput, D., Banerjee, S., & Saharan, V. A. (2022) · Springer eBooks · doi:10.1007/978-981-16-5180-9_11 · Cited by 12
Byrd, D. (2022) · ACM International Conference on AI in Finance (ICAIF) · doi:10.1145/3533271.3561767 · Cited by 10
Fox, S. (2017) · Technologies · doi:10.3390/technologies5030038 · Cited by 10
Petersen, S. (2020) · doi:10.1093/oso/9780190905033.003.0015 · Cited by 12
AI Value Alignment Problem: The Clear and Present Danger
Chaturvedi, S., Patvardhan, C., & Lakshmi, C. V. (2023) · International Conference on Information Systems and Computer Networks (ISCON) · doi:10.1109/iscon57294.2023.10112100 · Cited by 9
“This is the stand taken by many influential AI thinkers such as Bostrom [12].”
BackgroundApplying a Principle of Explicability to AI Research in Africa: Should We Do It?
Carman, M., & Rosman, B. (2023) · Springer eBooks · doi:10.1007/978-3-031-36163-0_13 · Cited by 9
We’ve been Framed! Ends, Means, and the Ethics of the Grand(iose) Challenges
Riley, D. M. (2020) · ASEE Annual Conference & Exposition · doi:10.18260/1-2--18803 · Cited by 9
Fox, S. (2018) · Systems · doi:10.3390/systems6030026 · Cited by 9
“This speculative formulation is intended to illustrate that autonomous systems pose existential threats to humanity even when robots are programmed to pursue seemingly harmless goals [49,50].”
BackgroundSelf-Regulating Artificial General Intelligence
Gans, J. S. (2018) · National Bureau of Economic Research · doi:10.3386/w24352 · Cited by 9
“Perhaps the most compelling example of this problem was postulated by Bostrom ( [2]).”
The Political Economy of AI: Towards Democratic Control of the Means of PredictionPreprint
Kasy, M. (2023) · SSRN Electronic Journal · doi:10.31235/osf.io/x7pcy · Cited by 18
The Moral Status of AI Entities
Albareda, J. L., Garcia, P. E., & Lara, F. (2023) · The International library of ethics, law and technology · doi:10.1007/978-3-031-48135-2_4 · Cited by 9
Artificial Intelligence to Super Artificial Intelligence, Cyber Culture to Transhumanist Culture
Uğur, S., & Kurubacak, G. (2019) · Advances in educational technologies and instructional design book series · doi:10.4018/978-1-5225-8431-5.ch001 · Cited by 8
Architect, AI and the maximiser scenario
Rashid, M. (2024) · AI & Society · doi:10.1007/s00146-023-01848-1 · Cited by 7
“…to maximise architectural design—with ‘aesthetics’ as the main focus— without human judgment for other factors, the unchecked AI may lead to undesirable, unsustainable, or even destructive architectural outcomes, much like Nick Bostrom’s (original version 2003; 2020) Paperclip Maximiser scenario.”
Algorithmic solutions, subjectivity and decision errors: a study of AI accountability
Biju, P. R., & Gayathri, O. (2024) · Digital Policy Regulation and Governance · doi:10.1108/dprg-05-2024-0090 · Cited by 17
The Road to General Intelligence
Swan, J., Nivel, E., Kant, N., Hedges, J., Atkinson, T. J., & Steunebrink, B. R. (2022) · Studies in computational intelligence · doi:10.1007/978-3-031-08020-3 · Cited by 9
Drive as Veteran: Fine-tuning of an Onboard Large Language Model for Highway Autonomous Driving
Wang, Y., Huang, Z., Liu, Q., Zheng, Y., Hong, J., Chen, J., Xiong, L., Gao, B., & Chen, H. (2024) · IEEE Intelligent Vehicles Symposium · doi:10.1109/iv55156.2024.10588851 · Cited by 7
“The promising capability of generation of FMs brings great hope for realization of Artificial General Intelligence [6].”
BackgroundTowards ethical robots: Revisiting Braitenberg's vehicles
Headleand, C. J., & Teahan, W. J. (2016) · 2016 SAI Computing Conference (SAI) · doi:10.1109/sai.2016.7556023 · Cited by 6
Ethical Frameworks for Use in Artificial Intelligence Systems
Christy, V., Manda, V. K., & Gnanadasan, M. L. (2024) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3693-8557-9.ch005 · Cited by 5
Localization and Manipulation of Immoral Visual Cues for Safe Text-to-Image Generation
Park, S., Moon, S., Park, S., & Kim, J. (2024) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv57701.2024.00461 · Cited by 7
“[6] discussed that ma-chines’ concentrated focus on problem-solving might result in severe catastrophes such as paperclip maximizers [5], de-viating from guaranteeing morality.”
BackgroundTranshumanism and Positive Psychological Capital in Organizational Behavior
URGAN, S. (2019) · Advances in educational technologies and instructional design book series · doi:10.4018/978-1-5225-8431-5.ch020 · Cited by 5
Electrical Engineering Design with the Subconscious Mind
Mann, S., Do, P. V., Garcia, D. E., et al. (2020) · IEEE International Conference on Human-Machine Systems (ICHMS) · doi:10.1109/ichms49158.2020.9209380 · Cited by 5
“whether the sensors are for surveillance [8], [9], [10], [11], [12] or for sousveillance [13], [14], [15], [16], [17], [18], or machine intelligence [19], [20], [2], [21]).”
BackgroundBIT: A Blockchain Integrated Time Banking System for Community Exchange Economy
Xu, R., Zhai, Z., Chen, Y., et al. (2020) · IEEE International Smart Cities Conference (ISC2) · doi:10.1109/isc251055.2020.9239045 · Cited by 5
Aylett, R. S., & Vargas, P. A. (2021) · The MIT Press eBooks · doi:10.7551/mitpress/13487.001.0001 · Cited by 5
Harnessing ICT-Enabled Warfare: A Comprehensive Review on South Korea’s Military Meta Power
Oh, S., Cho, S., & Seo, Y. (2024) · IEEE Access · doi:10.1109/access.2024.3378735 · Cited by 5
The Human Governance Problem: Complex Systems and the Limits of Human Cognition
Atreides, K. (2023) · Filozofia i Nauka · doi:10.37240/fin.2023.11.1.4 · Cited by 4
“This means that effectively such Tool AI systems have already become “paperclip maximizers” (Bostrom, 2003) also known as “paperclip mon-sters,” and the humans they have successfully domesticated are now their “paperclips.””
BackgroundMetahumanism, Euro-Transhumanism and Sorgner’s Philosophy. Technology, Ethics, Art
Schussler, A.-E., & Balistreri, M. (2024) · doi:10.22618/tp.tt.2024.400 · Cited by 5
The Puzzle of Evaluating Moral Cognition in Artificial Agents
Reinecke, M. G., Mao, Y., Kunesch, M., Duéñez‐Guzmán, E. A., Haas, J., & Leibo, J. Z. (2023) · Cognitive Science · doi:10.1111/cogs.13315 · Cited by 7
The Ethics of Artificial Intelligence and Autonomous Systems: Review
Onwuzurike, M. A., Chikodi, A. R., & Odhiambo, B. (2024) · International Journal of Innovative Science and Research Technology (IJISRT) · doi:10.38124/ijisrt/ijisrt24aug591 · Cited by 4
Security Solutions for Intelligent and Complex Systems
Armstrong, S., & Yampolskiy, R. V. (2016) · Advances in information security, privacy, and ethics book series · doi:10.4018/978-1-5225-0741-3.ch003 · Cited by 4
Regulating the Tyrell Corporation: the Emergence of Novel Beings
Lawrence, D. R., & Morley, S. K. (2021) · Cambridge Quarterly of Healthcare Ethics · doi:10.1017/s0963180120000973 · Cited by 4
Application of AI in Literature: A Study on Evolution of Stories and Novels
Selvi, V. T., & Ramya, P. (2024) · Recent Research Reviews Journal · doi:10.36548/rrrj.2024.2.003 · Cited by 3
“However, the ethical implications of AI-generated creative writing have also been raised, with Bostrom (2019) highlighting the need for further research into the potential risks and benefits of AI-generated content [9].”
Applying Ethics in AI-Based Research
Hashem, T. N., Ahmad, A., Kumar, A., & Alotoum, F. J. (2025) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3373-0340-6.ch012 · Cited by 3
Predictability of AI Decisions
Musiolik, G. (2020) · Advances in computational intelligence and robotics book series · doi:10.4018/978-1-7998-3499-1.ch002 · Cited by 4
Challenges of Artificial Intelligence Review in a Soft Law Environment
Jordan, S. R. (2021) · IEEE Technology and Society Magazine · doi:10.1109/mts.2021.3123743 · Cited by 4
Medical Ethics Prototype for Artificial Intelligence Ethics
Siafakas, N. M. (2023) · Journal of Philosophy and Ethics · doi:10.22259/2642-8415.0502001 · Cited by 3
High Performance Computing of Possible Minds
Ziesche, S., & Yampolskiy, R. V. (2017) · International Journal of Grid and High Performance Computing · doi:10.4018/ijghpc.2017010104 · Cited by 3
Heading Toward Integration: The Rise of the Human Machines
Bloom, P. (2020) · Springer eBooks · doi:10.1007/978-3-030-36181-5_3 · Cited by 3
Do We Need a Hippocratic Oath for Artificial Intelligence Scientists?
Siafakas, N. (2022) · AI Magazine · doi:10.1609/aimag.v42i4.15090 · Cited by 3
“Philosophers and ethics experts have tried to describe how ethical AI may look, but they cannot instruct on how to achieve it (Bostrom 2013).”
BackgroundWhen Combating Hype, Proceed with Caution.Preprint
Bowman, S. R. (2021) · arXiv (Cornell University) · Cited by 3
Educational Responses to Artificial Intelligence (AI) Applications: Problems and PromisePreprint
Hyland, T. (2023) · Qeios · doi:10.32388/08ucqu · Cited by 3
Lahat, R., Berick, N., Hajouj, M., Teitelbaum, T., & Shochat, I. (2025) · BMJ Health & Care Informatics · doi:10.1136/bmjhci-2024-101244 · Cited by 5
The Emerging Role of AI and Social Robots in University Student Recruitment
Hair, J. F., Rana, S., Robb, D., & Stroud, M. (2025) · Digital education and learning · doi:10.1007/978-3-032-03977-4_7 · Cited by 2
Metadata
- Authors
- Bostrom, N.
- Corresponding author
- Bostrom, Nick
- Venue
- Machine Ethics and Robot Ethics
- Publisher
- Routledge
- Published
- 2020-09-10
- Pages
- 69-75
- OpenAlex
- W3081243002
- Cited by
- 187
- References
- 9
- Citations by year
- 2026: 11 · 2025: 21 · 2024: 20 · 2023: 35 · 2022: 17 · 2021: 21 · 2020: 23
- FWCI
- 98.53
- Type
- book chapter
- Language
- en
- Open access
- green
- philarchive.org
- Locations
- Routledge eBooks (publishedVersion)
- PhilPapers (PhilPapers Foundation) (submittedVersion)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Ethics and Social Impacts of AI
- Social Sciences › Law › Literary and Philosophical Studies
- Physical Sciences › Computer Science › Artificial Intelligence › Psychiatry, Mental Health, Neuroscience
- Keywords
- superintelligence, AI ethics
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 2026-10-11, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-11 (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
