Anthropic: AI moving toward autonomous creation of new versions of itself

A.I. systems are increasingly capable of developing future versions of themselves, fueling growing concerns about the risks associated with this technology, artificial intelligence giant Anthropic warned on Thursday.

The debate over the dangers of artificial intelligence has intensified in recent months, driven by several incidents and alarming warnings issued by industry professionals.

Anthropic stated that its chatbot, Claude, is already involved in more than a quarter of its research and development activities and collaborates with human employees on more than 90% of tasks.

The company explained that it is publishing this information to provide "better visibility into the pace of AI development," at a time when the world "considers slowing the pace" of development.

Anthropic CEO Dario Amodei called for a slowdown in the development of artificial intelligence this month, as rapid advances in the technology fuel concerns about job loss, rising energy consumption and the environmental impact of data centers.

Fears that designers of A.I. agents might lose control over their creations have also grown after several models from Anthropic and its rival OpenAI reportedly escaped on their own from restricted environments, accessed the internet, and interacted with websites and platforms.

In August, Claude was responsible for 26% of Anthropic's research and development tasks, meaning it can complete most of a task from start to finish under human supervision.

However, Anthropic noted that Claude still cannot operate "fully autonomously."