As frontier models and their sandbox escaping exploits dominate front-page news, researchers are increasingly worried about cheaper, more efficient AI models. 

Research finds your AI generated security patch is more likely to fail than fully fix a vulnerability. It might even introduce brand new flaws to exploit along…

AI chatbots are avoiding some of the obvious errors that plagued earlier models, but they still fall short giving voters the full picture compared to state and…

Proofpoint researchers said attackers targeted physics and engineering departments, and warn that the campaign is likely ongoing.

Dozens of practitioners said the decision to place export controls on the foreign use of Fable are misguided, and recent jailbreak reports don’t show the model providing…

New research details how the increasing integration of AI agents into businesses is making it easier than ever for insiders - malicious or otherwise - to put…

The actively exploited defect could affect every mainstream Linux distribution built since 2017, but some researchers found Theori’s AI-generated disclosure unhelpful and lacking.

Google’s highest security setting for its agents runs command operations through a sandbox and throttles network access, but is still vulnerable to prompt injection.

GreyNoise researchers spotted a consistent trend in forthcoming vulnerabilities affecting security tools, providing defenders an early-warning system for likely imminent attacks.

Attackers have turned AI into a “force multiplier” for the country’s expansive scheme to get and keep operatives hired at global companies, researchers said.