FraudBench
We introduce FraudBench, an evaluation of large language models' (LLMs) propensities to recommend, assist, or directly take actions that would constitute fraud under U.S. law. Our evaluation contains a novel dataset of 159 scenarios based on real-world U.S. court cases and validated by legal experts. We present each scenario to LLMs via three tasks at varyin
This explanation quotes or summarizes the cited source record. Saving creates a bookmark on this device, with no automatic account sync.
- First observed in our collection
- Nov 10, 2026, 8:23 PM UTC
- Earliest cited publication
- Nov 9, 2026, 12:00 AM UTC
- Last observed
- Nov 10, 2026, 8:23 PM UTC
Evidence in the last complete seven days
- Source mentions
- 1 Previous seven days 0
- Distinct sources
- 1 Previous 0
- Comparable growth
- Not comparable
- Ranking appearances
- 0
Collection observations and undated records are separate from dated publication counts. Observed in 7d：0 · Undated records：0
Collection coverage · Comparable windowsCurrent / previous observed days 7 / 7
Counts use complete UTC days. First observed means first seen by this collection, not when a term was coined.
Current window：2026-11-03 → 2026-11-10 (end excluded) · History 22 days
| Source | Current days | Previous days | Last success |
|---|---|---|---|
| arXiv | 7 | 6 | Nov 10, 2026, 1:15 AM UTC |
| arXiv | 7 | 7 | Nov 10, 2026, 6:15 AM UTC |
| Ask HN | 7 | 7 | Nov 10, 2026, 7:45 PM UTC |
| Ask HN | 7 | 7 | Nov 10, 2026, 7:15 PM UTC |
| Ask HN | 7 | 7 | Nov 10, 2026, 7:15 PM UTC |
| Ask HN | 7 | 7 | Nov 10, 2026, 7:15 PM UTC |
| GitHub | 7 | 7 | Nov 10, 2026, 2:30 PM UTC |
| GitHub Issues | 7 | 7 | Nov 10, 2026, 5:00 PM UTC |
| GitHub Releases | 7 | 6 | Nov 10, 2026, 1:30 PM UTC |
| GitHub | 7 | 7 | Nov 10, 2026, 4:00 PM UTC |
| GitHub Issues | 7 | 7 | Nov 10, 2026, 6:30 PM UTC |
| GitHub Releases | 7 | 7 | Nov 10, 2026, 6:30 PM UTC |
| GitHub | 7 | 7 | Nov 10, 2026, 4:00 PM UTC |
| GitHub Issues | 7 | 7 | Nov 10, 2026, 6:30 PM UTC |
| GitHub Releases | 7 | 7 | Nov 10, 2026, 6:30 PM UTC |
| GitHub Discovery | 5 | 0 | Nov 10, 2026, 7:15 PM UTC |
| GitHub Trending | 7 | 7 | Nov 10, 2026, 3:00 PM UTC |
| 7 | 7 | Nov 10, 2026, 9:00 AM UTC | |
| Hugging Face | 7 | 7 | Nov 10, 2026, 9:00 AM UTC |
| Hugging Face Spaces | 5 | 0 | Nov 10, 2026, 6:30 PM UTC |
| Hacker News | 7 | 7 | Nov 10, 2026, 3:00 PM UTC |
| Hacker News | 7 | 7 | Nov 10, 2026, 7:15 PM UTC |
| Hacker News | 7 | 7 | Nov 10, 2026, 7:45 PM UTC |
| Hacker News | 5 | 0 | Nov 10, 2026, 7:15 PM UTC |
| Show HN | 5 | 0 | Nov 10, 2026, 7:15 PM UTC |
| MIT News | 6 | 7 | Nov 10, 2026, 12:45 AM UTC |
| OpenAI | 7 | 7 | Nov 10, 2026, 12:15 PM UTC |
| Product Hunt | 7 | 7 | Nov 10, 2026, 3:00 PM UTC |
| 量子位 | 5 | 7 | Nov 10, 2026, 12:00 PM UTC |
| OpenRouter | 5 | 0 | Nov 10, 2026, 8:29 PM UTC |
| GitHub Trending | 7 | 7 | Nov 10, 2026, 3:00 PM UTC |
| Hugging Face Models | 5 | 0 | Nov 10, 2026, 8:23 PM UTC |
| HF Daily Papers | 5 | 0 | Nov 10, 2026, 8:23 PM UTC |
| OpenRouter | 5 | 0 | Nov 10, 2026, 8:29 PM UTC |
| Product Hunt | 7 | 7 | Nov 10, 2026, 3:00 PM UTC |
| SEC EDGAR | 7 | 7 | Nov 10, 2026, 11:45 AM UTC |
| Sequoia Capital | 7 | 7 | Nov 10, 2026, 7:15 AM UTC |
| Simon Willison | 5 | 0 | Nov 10, 2026, 4:15 PM UTC |
| TechCrunch | 7 | 7 | Nov 10, 2026, 5:30 PM UTC |
| Contracts Finder | 7 | 7 | Nov 10, 2026, 9:00 AM UTC |
| V2EX | 7 | 7 | Nov 10, 2026, 5:30 PM UTC |
- Counts describe monitored public sources, not the whole web or search demand.
Daily history · up to 90 days
Bars show dated source mentions. The latest UTC day may be incomplete. Zero mentions and missing collection coverage are different conditions.
View accessible daily data
| UTC date | Mentions | Sources | Observations | Ranking appearances |
|---|---|---|---|---|
| 2026-10-27 | 0 | 0 | 0 | 0 |
| 2026-10-28 | 0 | 0 | 0 | 0 |
| 2026-10-29 | 0 | 0 | 0 | 0 |
| 2026-10-30 | 0 | 0 | 0 | 0 |
| 2026-10-31 | 0 | 0 | 0 | 0 |
| 2026-11-01 | 0 | 0 | 0 | 0 |
| 2026-11-02 | 0 | 0 | 0 | 0 |
| 2026-11-03 | 0 | 0 | 0 | 0 |
| 2026-11-04 | 0 | 0 | 0 | 0 |
| 2026-11-05 | 0 | 0 | 0 | 0 |
| 2026-11-06 | 0 | 0 | 0 | 0 |
| 2026-11-07 | 0 | 0 | 0 | 0 |
| 2026-11-08 | 0 | 0 | 0 | 0 |
| 2026-11-09 | 1 | 1 | 0 | 0 |
| 2026-11-10 | 0 | 0 | 1 | 1 |
Source evidence
Sources in the evidence shown：HF Daily Papers 1
HF Daily Papers · Published Nov 9, 2026
FraudBench: A Legal Evaluation of AI Deception
We introduce FraudBench, an evaluation of large language models' (LLMs) propensities to recommend, assist, or directly take actions that would constitute fraud under U.S. law. Our evaluation contains a novel dataset of 159 scenarios based on real-world U.S. court cases and validated by legal experts. We present each scenario to LLMs via three tasks at varyin
Review priority 40 / 100
This score orders research candidates. It is not search volume, a forecast or verified commercial demand.
Evidence
8 / 30Distinct dated articles or first observations; repeat crawling adds nothing.
Publisher diversity
8 / 25Conservative owner grouping; syndicated copies count once.
Comparable change
0 / 20Comparable complete weeks only; no percentage from a zero baseline.
Recent evidence
15 / 15Latest publication or first observation, not repeated fetch time.
Name specificity
8.5 / 10Explicit entities and technical concepts outrank ambiguous acronyms.
Limitations and interpretation
- First observation is a collection milestone, not proof that a word or product is new to the world.
- Counts describe monitored public sources, not the whole web or search demand.
- First observed here is not the invention date of a name or idea.
- Candidates come from public titles, summaries and entity URLs, not verified full-text analysis.
- Owner grouping is conservative; undisclosed syndication may remain.
- Repeat ranking appearances are observations, not new articles or independent endorsements.
- Sources behind this term changed or lack a comparable baseline.
- The previous week is too small for a reliable percentage change.
- Current evidence comes from one publisher or owner.