AI threats in the wild: The current state of prompt injections on the web

This Google Security Blog collection highlights multiple high-priority defenses and emerging threats across AI, endpoint, and cryptographic domains. Google flags Indirect Prompt Injection (IPI) as a top attack vector, reporting active web monitoring and continuous mitigations for Workspace/GenAI to harden LLMs and agentic workflows. For device security, Pixel teams are integrating a memory-safe Rust DNS parser into the baseband to reduce memory-safety vulnerabilities. Chrome and Google Account teams are deploying Device Bound Session Credentials (DBSC) to proactively prevent cookie/session-the