The trap Anthropic built for itself
Anthropic, OpenAI, Google DeepMind and others have long promised to govern themselves responsibly. Now, in the absence of rules, there's not a lot to protect them.
TL;DR
- The Trump administration blacklisted Anthropic, preventing it from doing business with the Pentagon.
- Anthropic refused to allow its AI to be used for mass surveillance or autonomous armed drones.
- Max Tegmark criticizes AI companies for resisting binding regulation, leading to their current issues.
- Companies like Anthropic, OpenAI, and Google DeepMind have broken safety pledges and resisted external rules.
- The lack of AI regulation is compared to having less oversight than for sandwich shops.
- The argument that regulation would cede AI dominance to China is challenged, as China itself is banning certain AI applications.
- Superintelligence is framed as a national security threat rather than an asset.
- AI progress has been rapid, with human-level language and knowledge mastery achieved faster than predicted.
- Tegmark believes treating AI companies like any other business, with mandatory safety trials, could lead to a beneficial AI future.
- Sam Altman of OpenAI stated he shares Anthropic's red lines regarding AI development.