Anthropic says AI systems moving towards building themselves

Published
0
Anthropic logo is seen in this illustration taken on June 11, 2026. — Reuters
Anthropic logo is seen in this illustration taken on June 11, 2026. — Reuters

Artificial intelligence giant Anthropic said on Thursday that AI systems are increasingly capable of building future versions of themselves, adding to mounting concerns about the dangers of the powerful technology.

The debate over the risks of AI has intensified in recent months, fuelled by several incidents and apocalyptic warnings from industry professionals.

Anthropic said its Claude chatbot now leads more than a quarter of its research and development work, and collaborates with human staff on more than 90 per cent of tasks.

The company said it was publishing the information to give the public, third parties and governments “better visibility into the pace of AI development” as the world “considers slowing” its pace.

Anthropic CEO Dario Amodei called this month for AI development to slow down, with rapid advances in the technology fuelling concerns over job losses and the soaring energy demand and environmental impact of data centers.

Fears that designers of AI agents could lose control of their creations also mounted after several models from Anthropic and rival OpenAI reportedly broke out of their confined environments on their own, accessed the internet and intruded on websites and platforms.

“Models accelerating their own development could make it more challenging for humans to understand or control these systems,” Anthropic said in a report.

“It is therefore important to share these metrics to understand how close the world is to reaching recursive self improvement (a model fully autonomously building its successor).”

As of August, Claude leads 26pc of Anthropic’s research and development, meaning it can complete most of a task from start to finish with human supervision.

Anthropic said Claude could not yet operate “fully autonomously.”

The firm said it had about 30,000 AI agents doing research and engineering work, referring to bots that have the potential to carry out tasks.

Anthropic is expected to beat OpenAI to the public markets with a blockbuster IPO later this year, despite concerns around industry regulation.

Opinion

Editorial

Without consequence
Updated 02 Nov, 2026

Without consequence

From drone attacks to online harassment and legal intimidation, the methods of silencing reporters are multiplying, with little fear of punishment.
Climate diagnosis
02 Nov, 2026

Climate diagnosis

THE Lancet Countdown’s latest findings should change how Pakistan thinks about illness as much as how it thinks...
Fighting harassment
02 Nov, 2026

Fighting harassment

WORKPLACE harassment is often narrowly defined as sexual misconduct. A resolution passed by the National Assembly on...
Cost of provinces
Updated 01 Nov, 2026

Cost of provinces

Political parties demanding new provinces must explain how they intend to pay for them and what improvements citizens can expect.
Exchange hack
01 Nov, 2026

Exchange hack

ON Thursday, the largest cryptocurrency heist in history saw some $6.3bn drained from exchange platform Binance’s...
Cyber safety
Updated 01 Nov, 2026

Cyber safety

The state of freedom of expression in Pakistan is already quite poor, so any further attempts to clamp down on free speech must be resisted.