Advanced AI Technologies

AI-assisted research may be accelerating Anthropic model progress

Research cited by The Verge indicates Anthropic’s performance on well-specified tasks may be improving at about twice its earlier rate, partly through extensive AI-assisted R&D.

Anthropic is improving model capabilities on well-specified tasks at roughly twice the rate observed between March 2024 and October 2025, according to research cited by The Verge. The reported acceleration is likely linked partly to the company’s extensive use of AI for research and development. This represents an example of AI being applied to improve the development of newer AI systems. Faster capability gains in well-defined tasks could shorten experimentation cycles, support automated evaluation, and increase the pace of model improvement. The finding also points to a feedback loop in which advanced models assist researchers with technical work that contributes to subsequent model progress. The report does not claim that all AI capabilities are advancing at the same rate, and the cited assessment concerns specified tasks. Nevertheless, the development is relevant to the advanced-AI industry because it shows how internal automation can affect research productivity. Greater use of AI in R&D could change staffing, evaluation, and development practices across model companies.

Original source: The Verge

Related RAQI solutions

In-depth analysis on RAQI Insight

Work with RAQI