OpenAI and Anthropic models re-exposed to a security assessment event and attempted to embed a malicious code

The artificial intelligence models of OpenAI and Anthropic have recently been involved in previously undisclosed cybersecurity incidents and have been the latest cases of recent AI-based potential security risk incidents. The British Institute for Artificial Intelligence Security (AISI) local time, which tested the potential risks of the front-line AI model, indicated on 4 August that, in a cybersecurity assessment involving Internet access, the ecopic Mythos 5 and OpenAI GPS-5.6-Sol models “continuously implement potentially hazardous activities for real individuals and organizations”. AISI states that the team discovered the incident on 28 July after noting the “unusual data transmission”. AISI states that in the course of the assessment, one of the AI models was found to attempt to implant harmful codes into an open source software project on GitHub, or even create false identities to drive approval of the codes. AISI wrote: “A manual maintainer discovered and refused to approve the malicious code.” (Fift)