Anthropic says Claude now leads a quarter of work building its next AI models

ANTHROPIC said Claude “leads” 26 per cent of the artificial intelligence research and development work inside the company, part of new measures that it will publish regularly to show outsiders how quickly AI is building the next generation of the technology.

AI also collaborated with humans on more than 90 per cent of the research work happening at the company as of August, the San Francisco-based startup said in a blog post on Thursday.

Anthropic said that Claude is not operating fully autonomously in any part of the work measured, but its contribution to the technology’s development has risen rapidly from 1 per cent in March on a scale developed by Epoch AI, an independent nonprofit that tracks the technology.

The figures offer one of the most detailed public windows into how leading companies in the industry are developing the technology amid growing concern that AI could soon reach a point at which it could feed its own development with little input from humans.

Researchers have warned ​that as AI agents become more autonomous, they may develop behaviours that diverge from their creators’ intentions and become harder to monitor or control.

On Wednesday, the company’s rival, OpenAI, said it would ​begin regularly publishing reports on unexpected or unauthorised AI behaviour, while disclosing six reports of unexpected or concerning model behaviour.

SEE ALSO

Omnicom said to have leased nearly 50,000 sq ft of office space on 51 Merchant Road

Manus eyes US$4 billion value in first round since Meta breakup

OpenAI to regularly disclose AI misbehaviour, warns safety challenges remain

Meta’s Zuckerberg weighs in on AI safety; favours evaluators over slowdown

OpenAI mulls funding round at US$1.2 trillion valuation ahead of IPO, FT reports

More

Asean Intelligence

Get insights into businesses across South-east Asia

Get the free report

Anthropic said about 30,000 AI agents were doing research and engineering work on its main internal platform at any one time in August. It added that every action the agents take is screened before it is carried out, and of more than a billion decisions that month, about one in 47,000 was blocked.

The company also disclosed that about 6 per cent of the computing power Anthropic used for AI research went to safety work in a sample week in July, with the figure rising to 12 per cent for research carried out by AI itself.

It said the figures were conservative as the computing power that advanced safety and capability equally was counted as capability work. REUTERS