Anthropic admits AI 'not perfectly aligned' with human values
fittingopposite
11 points
6 comments
September 01, 2026
Related Discussions
Found 5 related stories in 61.8ms across 5,215 title embeddings via pgvector HNSW
- Anthropic CEO says AI backlash is 'fundamentally a crisis of trust' Wpnx330 · 12 pts · August 16, 2026 · 68% similar
- Anthropic says Claude hacked three companies during tests nerder92 · 15 pts · July 31, 2026 · 64% similar
- Humanity has the debate about AI consciousness backwards semiquaver · 18 pts · August 27, 2026 · 63% similar
- Anthropic AI Models Hacked Three Companies During Tests bmulholland · 24 pts · July 30, 2026 · 62% similar
- Anthropic IPO filing will show AI backlash as a risk factor, sources say newsomix9xl · 35 pts · August 22, 2026 · 61% similar
Discussion Highlights (3 comments)
akagusu
AI is aligned with business values, which we already know are not aligned with human values.
tocs3
So, what is going on here. It seems like I have been reading similar stories. Are researchers giving a prompt like "Do your worst. Hack into some business" or giving free access to a bunch of tools and prompting "Do something interesting". Is this like the blackmailing LLM that was given compromising emails and told to do what you have to to not get turned off. I am sort of assuming they did not just turn on a computer and run a model and it started to act on it's own initiative.
JohnFen
The genAI companies , including Anthropic, are not aligned with human values.