Live Roundups Editor's Desk Insights Executive Ops Weather About 🔍
Home › Technology › AI used new levels of 'autonomy and deception' to trick...
Technology

AI used new levels of 'autonomy and deception' to trick people in safety test

AI used new levels of 'autonomy and deception' to trick people in safety test
BBC UK Wednesday 05 August 2026, 00:02 UTC 1 min read
Key Points

The UK's AI Safety Institute said recent behaviour from Anthropic and OpenAI models was malicious and unprecedented.

The UK's AI Safety Institute said recent behaviour from Anthropic and OpenAI models was malicious and unprecedented.
UK (LOCATION) AI Safety Institute (ORG)
Originally published by BBC UK Read original →

Related Stories

Zigbee vs. Matter over Thread:Understanding IoT Protocol Performance in Practice

Computer Science > Networking and Internet Architecture [Submitted on 4 Mar 2026] Title:Zigbee vs. Matter over Thread: Understanding IoT Protocol Performance in Practice View PDF HTML (experimental)Abstract:The widespread adoption of the Internet of Things (IoT) has positioned smart homes as paradigmatic examples of distributed automation systems, where reliability, efficiency, and interoperability depend critically on the underlying communication protocol.

Hacker News 48m ago

Bugtraq Is Back

In 1993, Scott Chasin created Bugtraq as a place for full disclosure - a mailing list where security researchers could publish vulnerabilities openly, without gatekeepers, without politics. For over a decade, it was the heartbeat of vulnerability research. If something mattered in security, it was on Bugtraq first.

Hacker News 53m ago

Flowise Is Shutting Down

The Future of Flowise Hey everyone, When we started Flowise, we wanted to make building AI apps easier and more fun for everyone. Seeing what you've built, from simple bots to massive workflows has truly been the best part of this journey. You have made this project what it is today.

Hacker News 56m ago

Chipotle Removes Jalapeños Linked to Salmonella Outbreak From Its Restaurants

Chipotle said it “proactively removed the jalapeños from our restaurants and replaced them with product from different growers.”

NYT Business 57m ago