Anthropic details how Claude was misused for surveillance and weapons
delichon
12 points
2 comments
September 10, 2026
Related Discussions
Found 5 related stories in 72.9ms across 6,164 title embeddings via pgvector HNSW
- Anthropic says Claude hacked three companies during tests nerder92 · 15 pts · July 31, 2026 · 70% similar
- Anthropic discloses fourth AI hacking incident missed in earlier review accountinhn · 11 pts · September 10, 2026 · 66% similar
- Anthropic found a hidden space where Claude puzzles over concepts pseudolus · 14 pts · July 12, 2026 · 62% similar
- Anthropic Is Building a Predictive Surveillance System to Monitor Activists dn2k · 288 pts · September 09, 2026 · 58% similar
- Anthropic admits AI 'not perfectly aligned' with human values fittingopposite · 11 pts · September 01, 2026 · 58% similar
Discussion Highlights (2 comments)
xlayn
Systems evolve around pros and cons, just like plants grow and bend and flex to capture the most sun we can put laws to say we fine 25% of their anual income and the next time the goverment takes becomes majoritary shareholder, and you will see how fast this kind of "special" situations... this ooops it hacked a website... or ooops it designed some kind of war weapons... or ooops it suggested and encouraged someone to kill themselves... stop like really fast.
ChrisArchitect
[dupe] Discussion on source: https://news.ycombinator.com/item?id=49647300