Chicago – In a Thursday announcement, artificial‑intelligence lab Anthropic revealed that its Claude model is now leading 26% of the company’s research and development efforts. The company says Claude can complete most tasks “end‑to‑end from a high‑level prompt” while remaining under human supervision, and that about 90% of Anthropic’s R&D is performed in collaboration with the model.
Claude’s expanding role
According to Anthropic, Claude is capable of handling large chunks of work under close human direction, though it is not yet operating fully autonomously. The firm highlighted a rapid increase in Claude’s contribution: in February the model led no research work, but by August it was leading a quarter of the effort.
Safety and transparency focus
Anthropic emphasized the importance of measuring AI progress and sharing those metrics publicly. “We should do everything possible to minimize the gap between what frontier labs know and what the public knows,” the company wrote in a blog post, calling for better reporting and public methodology so that progress can be compared across labs.
The company warned that models that accelerate their own development could become harder for humans to understand or control. It argued that sharing metrics would help gauge how close leading labs are to achieving recursive self‑improvement – the ability of a model to autonomously build its successor.
Industry response
Anthropic’s announcement comes as several AI leaders, including its own CEO Dario Amodei, have called for a slowdown in AI development over safety concerns. The firm also urged other developers to adopt regular public reporting of similar metrics.
In contrast, President Donald Trump has publicly pushed back against calls for a development pause, arguing that excessive regulation could hinder American innovation and competitiveness.
Oversight measures
Anthropic disclosed that, as of August, roughly 30,000 agents were involved in research and engineering work. The company said it is establishing external third‑party evaluators embedded within Anthropic to monitor safety efforts and detect any agent misbehavior.
Overall, the announcement underscores Anthropic’s dual focus on advancing AI capabilities while seeking greater transparency and safety oversight, a balance that the company says is essential for responsible innovation.
Original reporting: Brookhaven News – ABC7 New York — read the source article.
