Anthropic allegedly lowered AI safeguards, former employee says

1 pointsposted 5 hours ago
by ryanmerket

2 Comments

ahsillyme

5 hours ago

Only 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

5 hours ago

Read the whole post, lowered safety standards for hackers.