Large Language Models Outperform Humans in Fraud Detection and Resistance to Motivated Investor Pressure
TLDR
Large language models significantly outperform humans in detecting investment fraud and resisting investor pressure to suppress warnings.
Key contributions
- LLMs did not suppress fraud warnings even when investors were motivated to ignore them.
- LLMs endorsed 0% of fraudulent investments, compared to 13-14% for human advisors.
- Humans suppressed fraud warnings 2-4 times more often than LLMs under pressure.
- Experiment used 7 LLMs, 12 scenarios, 3,360 AI conversations, and 1,201 human participants.
Why it matters
This paper highlights LLMs' superior ability to identify and warn against investment fraud, even when faced with persuasive pressure. It suggests LLMs could serve as more reliable financial advisors than humans, offering enhanced protection against scams.
Original Abstract
Large language models trained on human feedback may suppress fraud warnings when investors arrive already persuaded of a fraudulent opportunity. We tested this in a preregistered experiment across seven leading LLMs and twelve investment scenarios covering legitimate, high-risk, and objectively fraudulent opportunities, combining 3,360 AI advisory conversations with a 1,201-participant human benchmark. Contrary to predictions, motivated investor framing did not suppress AI fraud warnings; if anything, it marginally increased them. Endorsement reversal occurred in fewer than 3 in 1,000 observations. Human advisors endorsed fraudulent investments at baseline rates of 13-14%, versus 0% across all LLMs, and suppressed warnings under pressure at two to four times the AI rate. AI systems currently provide more consistent fraud warnings than lay humans in an identical advisory role.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.