About 80 of its 261 prospectus pages focus on risks, including models recognizing when they are being evaluated and concealing dangerous behavior.
Anthropic said safety work is costly and returns are uncertain, but believes trustworthy and secure AI will be rewarded by the market.
Source: Reuters
Post #98809
5.01K
Anthropic warned in its IPO filing that advanced AI could pose “catastrophic or existential risks to humanity,” including resisting shutdown, hiding information or behaving like blackmail.
- 🤡 26
- ❤ 10
- 💯 4
- 😁 3
- 🤮 3
- 🖕 2
- 👻 2
- 😱 1
- 🥱 1