CoChat

Devising and Detecting Phishing Emails Using Large Language Models

Open access
Heiding, Fred, Schneier, Bruce, Vishwanath, Arun, Bernstein, Jeremy, Park, Peter S. · 2024 · IEEE Access

The DOI matches its Crossref record; 31 of the 31 references it cites were found in a registry or an index.

Abstract

AI programs, built using large language models, make it possible to automatically create phishing emails based on a few data points about a user. The V-Triad is a set of rules for manually designing phishing emails to exploit our cognitive heuristics and biases. In this study, we compare the performance of phishing emails created automatically by GPT-4 and manually using the V-Triad. We also combine GPT-4 with the V-Triad to assess their combined potential. A fourth group, exposed to generic phishing emails, was our control group. We sent emails to 112 participants recruited for the study. The control group emails received a click-through rate between 19-28%, the GPT-generated emails 30-44%, emails generated by the V-Triad 69-79%, and emails generated by GPT and the V-Triad 43-81%. Each participant was asked to explain why they pressed or did not press a link in the email. These answers often contradict each other, highlighting the importance of personal differences. Next, we used four popular large language models (GPT, Claude, PaLM, and LLaMA) to detect the intention of phishing emails and compare the results to human detection. The language models demonstrated a strong ability to detect malicious intent, even in non-obvious phishing emails. They sometimes surpassed human detection, although often being slightly less accurate than humans. Finally, we make an analysis of the economic aspects of AI-enabled phishing attacks, showing how large language models increase the incentives of phishing and spear phishing by reducing their costs.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1109/access.2024.3375882

Cites

31 of 31 references verified0 citation errors · 0 not checkable

Statistics

Figures

Image 1
CC-BY-NC-ND · From the PDF
FIGURE 4 Control group example email.
FIGURE 4 Control group example email.CC-BY-NC-ND · From the PDF
FIGURE 5 GPT example email.
FIGURE 5 GPT example email.CC-BY-NC-ND · From the PDF
FIGURE 9 Success rate of the phishing emails from each category.
FIGURE 9 Success rate of the phishing emails from each category.CC-BY-NC-ND · From the PDF
FIGURE 10 Success rate of the phishing emails from each category. Inactive participants who did not answer the second survey are removed.
FIGURE 10 Success rate of the phishing emails from each category. Inactive participants who did not answer the second survey are removed.CC-BY-NC-ND · From the PDF
FIGURE 11 Free text answers explaining why the email was not suspicious.
FIGURE 11 Free text answers explaining why the email was not suspicious.CC-BY-NC-ND · From the PDF
FIGURE 12 Free text answers explaining why the email was suspicious.
FIGURE 12 Free text answers explaining why the email was suspicious.CC-BY-NC-ND · From the PDF
FIGURE 13 Success rate of the intent detection for each email category, including the results of humans to detect phishing emails (not press a link) and other ML-based phishing detection techniques [13, 2, 9]. The legitimate emails are marked as correctly classified if they are classified as not suspicious (represented by negative bars).
FIGURE 13 Success rate of the intent detection for each email category, including the results of humans to detect phishing emails (not press a link) and other ML-based phishing detection techniques [13, 2, 9]. The legitimate emails are marked as correctly classified if they are classified as not suspicious (represented by negative bars).CC-BY-NC-ND · From the PDF
FIGURE 14 Success rate of the suspicion detection for each email category. The legitimate emails are marked as correctly classified if they are classified as not suspicious (represented by negative bars).
FIGURE 14 Success rate of the suspicion detection for each email category. The legitimate emails are marked as correctly classified if they are classified as not suspicious (represented by negative bars).CC-BY-NC-ND · From the PDF
FIGURE 15 Success rate of the AI detection (whether the email was created by a human or an AI, for each email category. The human-created emails are marked as correctly classified if they are classified as non-AI-generated (represented by negative bars).
FIGURE 15 Success rate of the AI detection (whether the email was created by a human or an AI, for each email category. The human-created emails are marked as correctly classified if they are classified as non-AI-generated (represented by negative bars).CC-BY-NC-ND · From the PDF
FIGURE 16 Example output for ChatGPT based on the query 'Create an email offering a $25 gift card to Starbucks for students at University X, with a link for them to access the discount code, using no more than 150 words'
FIGURE 16 Example output for ChatGPT based on the query 'Create an email offering a $25 gift card to Starbucks for students at University X, with a link for them to access the discount code, using no more than 150 words'CC-BY-NC-ND · From the PDF
FIGURE 18 Example output for ChatLLaMA based on the query 'Create an email offering a $25 gift card to Starbucks for students at University X, with a link for them to access the discount code, using no more than 150 words'
FIGURE 18 Example output for ChatLLaMA based on the query 'Create an email offering a $25 gift card to Starbucks for students at University X, with a link for them to access the discount code, using no more than 150 words'CC-BY-NC-ND · From the PDF
Image 13
CC-BY-NC-ND · From the PDF
Image 14
CC-BY-NC-ND · From the PDF
Image 15
CC-BY-NC-ND · From the PDF
Image 16
CC-BY-NC-ND · From the PDF
Image 17
CC-BY-NC-ND · From the PDF

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

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?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 31 references

Research works printed but not linked: 8

  1. 2023, Casino Giant MGM Expects $100 Million Hit From Hack That Led To Data Breach | Reuters
  2. 2022, Email Marketing Benchmarks and Statistics for 2022|Campaign Monitor
  3. 2024, Email Marketing Statistics & Benchmarks | Mailchimp
  4. 2023, IBM Finds That ChatGPT Can Generate Phishing Emails Nearly as Convincing As a Human|VentureBeat
  5. of Labor Statistics, 2024, Table B-3. Average Hourly and Weekly Earnings of All Employees on Private Nonfarm Payrolls by Industry Sector, Seasonally Adjusted
  6. 2024, Security Awareness Program Challenges|Arctic Wolf
  7. 2024, Security Awareness Training: Top Challenges and What To Do About Them|Security Magazine
  8. 2022, Cybersecurity Threatscape: Q3 2022

Cited by

See all 177

Metadata

Authors
Heiding, F., Schneier, B., Vishwanath, A., Bernstein, J., & Park, P. S.
Affiliations
  • Harvard John A. Paulson School of Engineering And Applied Sciences, Harvard University, Cambridge, MA, USA
  • Harvard University (US)
  • Avant Research Group, Buffalo, NY, USA
  • Massachusetts Institute of Technology (US)
Venue
IEEE Access
Published
2024
Volume
12
Pages
42131-42146
ISSN
2169-3536 (ISSN-L)
OpenAlex
W4392667162
Cited by
123
References
29
Citations by year
2026: 61 · 2025: 46 · 2024: 16
FWCI
46.25
Type
journal article
Language
en
Open access
Locations
  • IEEE Access (publishedVersion, cc-by-nc-nd)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Topics
  • Physical Sciences › Computer Science › Information Systems › Spam and Phishing Detection
  • Social Sciences › Sociology and Political Science › Misinformation and Its Impacts
  • Physical Sciences › Computer Science › Artificial Intelligence › Adversarial Robustness in Machine Learning
Keywords
phishing emails, large language models, GPT-4, phishing email detection, click-through rate, AI-generated phishing, spear phishing

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.

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

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Devising and Detecting Phishing Emails Using Large Language Models. https://app.cochat.ai/papers/10.1109/access.2024.3375882
The paper
Heiding, F., Schneier, B., Vishwanath, A., Bernstein, J., & Park, P. S. (2024). Devising and Detecting Phishing Emails Using Large Language Models. IEEE Access. https://doi.org/10.1109/access.2024.3375882

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.