How Gemini 3.1 Pro Helped a Redditor Expose a Dangerous Fraudster at His Gym
Procesius · reddit · 2026-10-27
A Reddit user recounts using Gemini 3.1 Pro's reasoning instincts to investigate a seemingly friendly 68-year-old man at his gym, ultimately uncovering that he was a dangerous fraudster linked to a man's death. A dramatic real-life case of AI-assisted detective work.
More from Fun
- Investors flooded with pitches from people who 'invented algorithms' via one ChatGPT chat — alliekmiller · 2026-10-27
- Google's audio model IDs a song in a noisy café where Shazam fails — keenanisalive · 2026-10-27
- OpenAI and Claude Both Claim 2^o(n) 3-SAT Algorithms, Refuting ETH — pdkj · 2026-10-27
- Andrew Carr jokes: post-AGI human activity will just be duels at dawn — andrew_n_carr · 2026-10-27
- Mathematicians clash over AI proof hype: 'bigger than Newton' claim called absurd — LucaAmb · 2026-10-27
- Professor spots ChatGPT-written 'personalized' PhD outreach emails instantly — tallinzen · 2026-10-27