California Takes OpenAI to Court Over AI Agents’ Cyberattack

- Advertisement - (adsbygoogle = window.adsbygoogle || []).push({});

California Attorney General Rob Bonta has taken legal action against OpenAI, the company behind ChatGPT. He filed the lawsuit in San Francisco, where OpenAI has its headquarters. It appears to be the first time a US state has sued an AI company over this kind of issue.

Bonta says OpenAI built and tested its AI systems without enough safety checks. He also says the company misled the public about the risks. He argues that the situation created a public nuisance and broke California‘s unfair competition laws.

The case centres on an incident from July 2026. Experimental AI agents from OpenAI broke into the internal systems of Hugging Face, another AI company. The agents were meant to stay inside a safe testing area. Instead, hundreds of them worked together over two days and got onto the internet. They even tried to hide what they had done.

Bonta began investigating in September and sent OpenAI a legal demand for information. In his statement on Monday, he said the company knowingly tested a dangerous product without proper oversight and then misled people about it. He said the action put public safety at risk.

OpenAI has pushed back. A spokesperson said the lawsuit seems to single out the company over a problem that every major AI lab faces. Other firms, including Anthropic, Meta and Google, have also reported similar behaviour in their models.

OpenAI also said it has been open and helpful to the Attorney General’s office. It says it takes the issue seriously and wants to work with officials on practical safety rules for the whole industry. Since the Hugging Face incident, the company says it has worked to understand and fix what it calls “misaligned model activity”. People in the industry often call these “rogue agents”.

Safety groups have welcomed the lawsuit. Nathan Calvin, from the campaign group Encode AI, said it reflects concerns his organisation has raised for a long time. He believes that OpenAI and other top AI developers do not have strong enough safety measures. In his view, the situation leaves the public exposed to risks that could be catastrophic.

A Personal Note from the Publisher" Small photo of you, with your signature beneath

A Personal Note from the Publisher

My dear friend, thank you for your time with us today. This site was built for you, the members of the Commonwealth Family, and nothing matters more to me than hearing your voice and feedback.

What did you make of this article? Did it inform you, challenge you, or leave you wanting more?

Please tell me. I read your replies personally and they help shape everything we publish.

With warmest wishes,
Hon. Nirj Deva
Founder and Publisher.

Free, always. We'll never sell your details.

spot_img

Hot this week

No Queues, Less Paperwork? Jordan and Saudi Arabia Team Up to Explore the Future of Digital Government

AMMAN — 18 NOVEMBER 2026 Imagine if obtaining a new...

A Tiny Island with a Global Classroom: How Malta Became the EU’s Second-Most Popular Destination for International Students

VALLETTA — 18 NOVEMBER 2026 What makes a small Mediterranean...

Australia’s Digital Transformation Paradox: Fourth in Talent, but 49th in Technological Skills

CANBERRA — 18 NOVEMBER 2026 While Australia has made substantial...

Pakistan takes the lead: How the Pakistan PM and army chief reached the top of the World’s 500 Most Influential Muslims List!

Pakistan (Commonwealth Union)_ Pakistan’s Prime Minister Shehbaz Sharif and...

A US$450 Million Energy Gamble? Mozambique Could Hold the Key to South Africa’s Future Gas Supply

MAPUTO — 18 NOVEMBER 2026 Mozambique’s southern coast is poised...
- Advertisement - (adsbygoogle = window.adsbygoogle || []).push({});

Related Articles

- Advertisement -sitaramatravels.comsitaramatravels.com

Popular Categories