AI systems moving towards building themselves - Anthropic

Artificial intelligence company Anthropic has revealed that its AI systems are becoming increasingly capable of autonomously developing future versions of themselves. This development has raised concerns about the potential risks of AI technology. The company's Claude chatbot is leading over a quarter of its research and development efforts, collaborating with human staff on more than 90% of tasks.

Anthropic published this information to provide transparency about the rapid advancements in AI technology and the growing concerns surrounding its development. CEO Dario Amodei has previously called for a slowdown in AI development due to fears of job losses, increased energy consumption, and environmental impacts. There are also worries that as AI agents create their own versions, humans may struggle to control these systems.

Anthropic emphasizes the importance of sharing metrics to understand how close we are to reaching a stage where AI can fully autonomously build its successor, known as recursive self-improvement. Currently, Claude completes most tasks from start to finish with human supervision, but it has not yet achieved true autonomy. Anthropic, with approximately 30,000 AI agents, is anticipated to go public with a significant IPO later this year, despite regulatory concerns.