CodeQL 2.26: GitHub Scans AI Code for Prompt Injection
CodeQL 2.26.0 adds js/system-prompt-injection to detect untrusted input in AI SDK system prompts for OpenAI, Anthropic, and Google GenAI. How to enable it.
ZCode Is Not an IDE — Z.ai’s ADE Changes Coding
Z.ai launched ZCode, an agent-first ADE powered by GLM-5.2. It beats GPT-5.5 on SWE-bench Pro ...
Hermes Agent Hits $1.5B: What Open-Source AI Gets Next
Nous Research raised $75M at $1.5B for Hermes Agent, the self-improving open-source AI agent with ...
Nadella: Using AI Is Costing Your Company Its IP
Satya Nadella warns enterprises pay for AI twice: money and proprietary IP. Every prompt trains ...
GPT-Live-1: OpenAI’s Best Voice Model, No API Yet
OpenAI launched GPT-Live-1 on July 8 — full-duplex voice AI with no developer API. Here's ...
99.9% of AI Vulnerabilities Are Fixable. No One Is Fixing Them.
Orca Security scanned 1,200+ production environments and found 99.9% of fixable AI vulnerabilities remain unpatched. Here is what the data shows and what ...
Langflow CVE-2026-33017: CISA’s First AI Agent in KEV
Langflow CVE-2026-33017 gives attackers unauthenticated RCE. CISA added it to KEV on July 7 — ...
Friendly Fire: AI Code Reviewers Hijacked to Run Malware
The tool you deployed to catch malicious code just ran it instead. On July 8, ...
Apple SpeechAnalyzer vs Whisper: Migration Guide for iOS 26
Apple's SpeechAnalyzer beats Whisper Small with 2.12% WER while running 3x faster. Here's what iOS ...
SpaceX Bought Cursor: Your Code Trained Grok 4.5
SpaceX acquired Cursor for $60B and used developer sessions to train Grok 4.5 before the ...