AI news story
We’re Less Safe From Cyber Risks Now, Says HackerOne CEO
Anthropic’s new model Mythos is raising fear among regulators and companies that AI could exploit cyber risks, even as a select group of companies test out the models to identify vulnerabilities. Kara Sprague, HackerOne CEO, discusses the state of cy
Editor's take
Anthropic's "Mythos" large language model is prompting concerns among cybersecurity professionals and regulators about its potential to amplify existing digital threats.
The core worry centers on the possibility that advanced AI, like Mythos, could be leveraged by malicious actors to discover and exploit system vulnerabilities at an unprecedented scale and speed. This development underscores a growing tension between the rapid advancement of generative AI capabilities and the current state of cybersecurity defenses, which may not be adequately prepared for such potent new attack vectors.
Future developments to monitor include how quickly security firms can develop AI-powered defenses to counter these AI-driven threats, and whether regulatory bodies will propose concrete measures to mitigate the risks posed by models capable of sophisticated vulnerability discovery. A key indicator of progress would be the emergence of AI tools specifically designed to audit and secure other AI systems before they are widely deployed.
Signal score: 5
This event was corroborated by 16 independent sources. The signal score weighs cross-source corroboration, recency, source weight and topic salience. How we rank stories.
Original reporting
This story summarises reporting published by Bloomberg. Read the original article at Bloomberg.