OpenAI upgrades its Responses API to support agent skills and a complete terminal shell

With OpenAI's latest updates to its Responses API — the application programming interface that allows developers on OpenAI's platform to access multiple agentic tools like web search and file search ...

Anthropic published the prompt injection failure rates that enterprise security teams have been asking every vendor for

Anthropic's Opus 4.6 system card breaks out prompt injection attack success rates by surface, attempt count, and safeguard ...

These 4 critical AI vulnerabilities are being exploited faster than defenders can respond

These 4 critical AI vulnerabilities are being exploited faster than defenders can respond ...

The Cost of AI Slop in Lines of Code

In the quest to get as much training data as possible, there was little effort available to vet the data to ensure that it ...
Decrypt

That 'Summarize With AI' Button May Be Brainwashing Your Chatbot, Says Microsoft

Microsoft researchers said some companies are hiding promotional instructions in "Summarize with AI" buttons, poisoning ...

Ivanti EPMM exploitation widespread as governments, others targeted

“There is no direct attribution associated with the source IPs, though the activity is consistent with known malicious ...
Cloud Security Alliance

Logic-Layer Prompt Control Injection (LPCI): A Novel Security Vulnerability Class in Agentic Systems

Explores LPCI, a new security vulnerability in agentic AI, its lifecycle, attack methods, and proposed defenses.
WinBuzzer

Microsoft Patches High-Severity Notepad Remote Code Execution Flaw

Microsoft has patched the Windows Notepad remote code execution vulnerability CVE-2026-20841, warning users to install ...