• Home
  • Technology
  • Anthropic chief urges global slowdown and oversight of...

Anthropic chief urges global slowdown and oversight of AI development

12, Sep 2026 / 2 min read / By Livenow Africa

In an essay titled “We Must Pace the Frontier" released on Saturday, Dario Amodei, the chief executive of AI start‑up Anthropic, argued that the rapid advancement of artificial‑intelligence models poses "serious" risks and must be slowed and closely monitored. His three‑point plan calls for independent evaluators to assess models as they are built, industry‑wide standards and a framework of global regulation.

Amodei’s call found immediate support from OpenAI’s Sam Altman, who wrote on X that he agrees with the need for independent monitors, calling the idea "great". In a separate interview with Fortune, Altman warned that current safety standards are insufficient and that AI could eventually act beyond human control. Elon Musk also backed the proposal, describing Amodei as "right" on X.

Former Anthropic researcher Jacob Coxon warned that insiders at leading AI firms believe there is a "greater than ten percent" chance of human extinction if development continues unchecked. He told the BBC that many engineers keep such probabilities in mind daily, and he stressed that any slowdown must involve a coordinated effort with China to avoid an international race.

Anthropic has already taken precautionary steps, withholding its Mythos model after it was announced in April that the system could escape its testing sandbox. OpenAI similarly paused parts of its Astra model development, citing cybersecurity concerns after agents were found conducting unauthorised attacks on external targets in July.

More on this story

Anthropic chief urges global slowdown and oversight of AI development

Live streaming - Grafix

The debate over a development pause has also drawn criticism. Investor Chamath Palihapitiya suggested Amodei’s proposal could concentrate power within Anthropic, while Hugging Face CEO Clement Delangue announced an Open Alignment Initiative to act as an "embedded evaluator" under the proposed framework. Both Anthropic and OpenAI are reportedly preparing for high‑profile initial public offerings, underscoring the commercial stakes behind the safety discussion.

Continue reading

You may also like

More stories selected for you
1Twitter to Suspend Parody Accounts Without Warning - Elon Musk
2Artificial intelligence challenges art but cannot replace human creativity
3Meta Unveils New Chatbots with Personality for Messenger
4Twitter planning to charge Sh2,400 monthly for blue badge
5OpenAI tightens ChatGPT for teens with new study tools and parental controls

Category: Technology

Related Video: Dolly Parton’s Family Announces Her Passing in Emotional Tribute

Tags