2 ms·
Anthropic allegedly lowered AI safeguards, former employee says
- ahsillyme 3mo agoOnly way to improve safety of software is the ability to find the flaws of said software. If they did lower safeguards, it's probably positive for software safety standards.
- ryanmerket 3mo agoRead the whole post, lowered safety standards for hackers.