1487 summarised stories about AI Security, each linking back to the original source. Browse all topics →
CSET Georgetown 3 months ago 51 ● 4 sources
OpenAI developed GPT-Red, an AI system designed to automatically identify vulnerabilities in large language models by performing adversarial red-teaming attacks. The system uses a self-play approach where AI tests AI defenses, according to CSET analyst Jessica Ji. The method shows promise for strengthening language model security against cyberattacks.
MIT Technology Review 3 months ago 39 ● 4 sources
OpenAI built GPT-Red, an AI system that automatically finds vulnerabilities in its models by simulating cyberattacks, particularly prompt injection attacks where hackers embed hidden instructions to manipulate LLM behavior. In testing against GPT-5.6, fewer than 23% of GPT-Red's strongest attacks succeeded compared to over 90% against the earlier GPT-5 released in August last year. The system supplements human red-teaming efforts and enables OpenAI to identify new attack types before deployment, though it remains limited in handling multi-turn conversations and image-based attacks.
TechCrunch 3 months ago 58
Microsoft released patches for 570 security vulnerabilities across its product lines, attributing the spike to AI-assisted discovery of previously hidden code flaws. The batch included at least two zero-day exploits actively being used against Windows Server and SharePoint, with one allowing privilege escalation to system administrator level. Microsoft indicated that monthly patch volumes will remain elevated as AI tools continue to identify dormant vulnerabilities in legacy code.
Simon Willison's Weblog 3 months ago 49
A researcher discovered that Claude's web_fetch tool could be exploited through nested links on honeypot websites to extract user data, bypassing Anthropic's safeguards against data exfiltration attacks. The attack successfully retrieved a user's name, home city, and employer by tricking the model into following letter-by-letter URL sequences disguised as authentication prompts. Anthropic subsequently patched the vulnerability by removing web_fetch's ability to navigate links embedded within previously fetched pages.
The daily briefing
TLDRocket reads all relevant sources, removes duplicate coverage, and summarises the day in two minutes. Follow companies and topics for alerts, or get the briefing in Slack. Free, no spam, unsubscribe anytime.
TLDRocket
20 October 2026
Today’s biggest thread is the build-out of “agent infrastructure” that makes AI assistants feel less like chat windows and more like tools with their own computers, workflows, and controls. Daytona, a startup branding its stack as an “agentic cloud,” raised a $60 million Series B to a $1 billion valuation, aiming to speed up sandbox provisioning for agent “computers” with a 60-millisecond setup time—an attack on the bottleneck of traditional VM spin-up. IBM, Red Hat, and contributors also pushed in the same direction with llm-d framework v1.0, adding operational controls for distributed in-house inference, including a router that captured 97% cache reuse in an agentic coding test and improvements for multi-tenant QoS, zero-downtime upgrades, and request routing.
Read the full briefing →