Your AI agent’s system prompt is not a security control

An AI agent told in its system prompt to show a user only what that user is cleared to see will hand over more the moment someone talks it into doing so. Gee Rittenhouse, who oversees Security Hub, GuardDuty, and Inspector at AWS, and Eric Johnson, a f… Continue reading Your AI agent’s system prompt is not a security control→

Windows memory integrity switches on automatically for eligible devices in October 2026

Beginning in October 2026, Windows quality updates start enabling memory integrity protection on eligible devices with little or no additional configuration. On machines where Virtualization-based Security is not already running, those same updates ena… Continue reading Windows memory integrity switches on automatically for eligible devices in October 2026→

Anthropic’s Enterprise Frontier Safeguards lets your Claude logs stay in your cloud

Eight members of the Analysis and Resilience Center for Systemic Risk, a group whose roster includes the CISOs of Goldman Sachs, Morgan Stanley, Citi, Bank of America, and Wells Fargo, spent months working with Anthropic on a question their examiners c… Continue reading Anthropic’s Enterprise Frontier Safeguards lets your Claude logs stay in your cloud→

An AI CAPTCHA solver talked itself out of the right answer

You have probably spent a few seconds of your life turning a picture until it lines up. Some sites, instead of asking you to tick a box, show you a circular chunk of a photo that has been spun around, and you drag it until the inside matches the ring a… Continue reading An AI CAPTCHA solver talked itself out of the right answer→

Bot detection arrives in CrowdSec 1.8.0, along with two DoS fixes

Failed SSH logins pile up in an auth log, and a scanner walks a website looking for exposed admin paths. CrowdSec reads log sources and HTTP requests, works out which addresses are misbehaving, and hands the block to a separate remediation component si… Continue reading Bot detection arrives in CrowdSec 1.8.0, along with two DoS fixes→

Debian developers rejected an LLM ban and left disclosure voluntary

A maintainer reading a merge request can’t tell whether a person or a model wrote the diff, and nobody has to say. Debian developers voted on that through August 28, and Kurt Roeckx, the project secretary, announced the result: the winning option… Continue reading Debian developers rejected an LLM ban and left disclosure voluntary→

Free ChatGPT users get ads picked from whatever they just asked about

Say you’re on the free plan and you ask ChatGPT to help you pick a mattress. An ad may turn up next to the answer, and it got there because of what you just asked about, plus your rough location and whatever device you’re on. What ChatGPT r… Continue reading Free ChatGPT users get ads picked from whatever they just asked about→