Anthropic warns it's getting closer to self-improving AI

Anthropic warns in a revised S-1A filing that it is nearing recursive self-improvement capability, noting that using AI in research can drive unexpected leaps in model capability.

The filing outlines safety considerations and policy updates aimed at keeping human supervision central to future development.

According to the document, which was obtained by journalist Julie Bohr, the company is already using artificial intelligence in research and development. However, it warns that the practice could accelerate progress at a faster pace and poses security risks.

Anthropic said it has previously used AI to develop new models and seen sharp jumps in their capabilities that it failed to recognize. At the same time, its internal estimates suggest it may soon reach a critical internal threshold of dramatically accelerating progress thanks to AI, while the financials in the document show revenues growing alongside expenses.

The evaluation of boundaries and the role of people

However, the company clarified that none of its current models have yet exceeded this threshold and is continuing to evaluate the data.

↳

READ ALSO

HACKS.GR / RELATED

AI Unlocked

How much is artificial intelligence threatened by high interest rates and global debt?

↗

Anthropic has pledged that humans will remain in the research and development process for the foreseeable future. The company does not expect to be able to safely use its systems to design and build the next generation of models fully autonomously without significant human supervision, and it is considering changes to its safety rules to keep humans in control.

>Stay connected with Hacks.gr

Set us as a preferred source on Google and find us on social media.

Preferred source

↗

end of article

Read the original English article on Hacks.gr