Anthropic’s new AI model (Claude) will scheme and even blackmail to avoid getting shut down
1 min read
Advertisements
Image credit: AI generated by Satheesh Sankaran via Pixabay. Article by Mihai Andrei. ZME Science – May 23, 2025.
In a simulated workplace test, Claude Opus 4 — the most advanced language model from AI company Anthropic — read through a stack of fictional emails. The test scenario was that Claude served as an assistant at a tech company and the AI discovered it was about to be deactivated and replaced with a newer system. But buried in those emails was a secret: the engineer responsible for shutting Claude down was having an extramarital affair. […]
Click here to view original web page at www.zmescience.com
Advertisements
GOOG
IBM
AMZN
AAPL
DELL
INTC
MSFT
TSLA
META
EBAY
NVDA
NFLX
HPE
LUMN
ORCL
Actually no matter if someone doesn’t be aware of then its up to
other people that they will assist, so here it happens.