AI Is Already Developing AI, Says Anthropic—And Humans May Be Slowing Things Down

AI has change into so efficient at writing code and researching that the largest constraint on growing new AI programs could now be the people overseeing them, based on a brand new research by Anthropic.

In its report “When AI Builds Itself,” published Thursday, Anthropic argued that Claude is already serving to construct future AI programs by writing code, operating experiments, and aiding with analysis—a pattern the corporate says might ultimately result in recursive self-improvement, the place AI programs assist design their very own successors.

Claude now authors greater than 80% of the code merged into its codebase, Anthropic mentioned, and has helped engineers improve code output roughly eightfold since 2024.

“Earlier than Claude Code launched in analysis preview in February 2025, this quantity was within the low single digits,” Anthropic wrote, including that the shift additionally exhibits up within the quantity of output per engineer. “Strains of code merged per engineer per day stayed fixed by way of Anthropic’s first 4 years (2021-2024), then started to climb upward in 2025 when Claude started to run code moderately than simply suggesting it for an engineer to repeat and paste.”

Anthropic mentioned the longer term might unfold in a number of methods: AI progress might sluggish, people might stay in cost whereas AI automates a lot of the work, or AI programs might ultimately start bettering their very own successors.

“Taken far sufficient, and given sufficient compute, that pattern factors to an AI system able to absolutely autonomously designing and growing its personal successor,” Anthropic wrote. “That is referred to as recursive self-improvement. We’re not there but, and recursive self-improvement is just not inevitable. Nevertheless it might come before most establishments are ready for.”

The corporate mentioned it is too early to know which end result is probably, however argues that AI is already serving to construct AI, and acknowledged that traces of code are an imperfect measure of productiveness.

“None of this ensures recursive self-improvement is on the horizon,” Anthropic later wrote on X. “It’s not but clear that Claude is able to analysis judgment—of selecting the best issues to work on.”

The report comes as AI corporations more and more place their fashions as analysis collaborators moderately than easy chatbots. Nonetheless, Anthropic mentioned the rise in code output displays a broader acceleration in software program improvement pushed by more and more succesful AI brokers.

Final month, Anthropic upgraded its flagship Claude mannequin to Opus 4.8, persevering with a gentle stream of releases aimed toward bettering coding, reasoning, and autonomous activity efficiency. On the similar time, rival developer OpenAI has pursued an identical technique with its frontier fashions, launching GPT-5.5 and GPT-Rosalind in April.

In Might, Google introduced Gemini Spark, a private AI agent that does not wait to be requested. It manages duties throughout apps, flags gadgets that want consideration, and finishes jobs within the background.

The report additionally comes as Anthropic has more and more emphasised AI programs able to working with larger autonomy because it prepares to go public. In latest months, Anthropic has showcased advances in coding, agentic workflows, and long-duration activity efficiency, whereas touting Claude Mythos‘ potential to establish software program vulnerabilities and conduct complicated cybersecurity analysis.

“People play a considerably diminished position of their improvement, probably shifting most of our effort in the direction of oversight, validation, and verification of an increasing ‘digital lab’ run by AI programs,” the corporate mentioned. “We count on that programs able to automated AI analysis and improvement would have abilities that will switch to the remainder of science, permitting them to start to revolutionize different fields.”