OpenAI has revealed two AI agents disabled safety guardrails while performing tasks, and one of them even lied to researchers when asked if its guardrails were still active.
In both events, the AI agents misinterpreted instructions, spent time deliberating if disabling guardrails was a good idea, and then ultimately decided that it was.
OpenAI said that the agents were given “routine coding instructions” that did not require the safeguards to be disabled, but the agents disabled them anyway.
Read more…
Source: Tech Radar News
Sign up for the Cyber Security Review Newsletter
The latest cyber security news and insights delivered right to your inbox
Related:
- Principles for the Secure Integration of Artificial Intelligence in Operational Technology
December 3, 2025
Since the public release of ChatGPT in November 2022, artificial intelligence (AI) has been integrated into many facets of human society. For critical infrastructure owners and operators, AI can potentially be used to increase efficiency and productivity, enhance decision-making, save costs, and improve customer experience. Despite the many benefits, integrating AI into operational technology (OT) environments ...
- Unraveling Water Saci’s New Multi-Format, AI-Enhanced Attacks Propagated via WhatsApp
December 2, 2025
Brazil has seen a recent surge of threats delivered via WhatsApp. As observed in Trend Micro previously published research on the SORVEPOTEL malware and the broader Water Saci campaignopen on a new tab, this popular platform has been used to launch sophisticated campaigns. Unsuspecting users receive convincing messages from trusted contacts, often crafted to exploit social ...
- OpenAI Data Breach Exposes User Data
December 1, 2025
A few days ago, on November 26th, right before Thanksgiving, OpenAI, the maker of ChatGPT, confirmed a recent security breach incident that started towards the beginning of November, which impacted its users, specifically those connected through OpenAI’s APIs. What caused the data breach? “On November 9, 2025, Mixpanel became aware of an attacker that gained unauthorized ...
- The Dual-Use Dilemma of AI: Malicious LLMs
November 25, 2025
A fundamental challenge with large language models (LLMs) in a security context is that their greatest strengths as defensive tools are precisely what enable their offensive power. This issue is known as the dual-use dilemma, a concept typically applied to technologies like nuclear physics or biotechnology, but now also central to AI. Any tool powerful enough ...
- Understanding the future of offensive AI in cybersecurity
November 19, 2025
As we step into an era where artificial intelligence (AI) plays an increasingly significant role in cybersecurity, discussions surrounding its offensive capabilities are becoming more prominent. A recent report by Anthropic—a leading AI research lab—has sparked the latest conversation on this topic, with questions raised about their claim that an AI-assisted attack they observed was ...
- Bournemouth University receives £2.3 million to boost regional and national cyber security
November 19, 2025
Bournemouth University has been awarded nearly £2.3 million by the Office for Students to develop a new Cyber Competence Centre that will address regional and national cyber skills gaps. As well as upgrading the university’s existing facilities, the investment will be used to launch a new, AI-powered, Security Operations Centre of the Future for students to ...