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

2026-05-10T13:23:37Z3189ee2abb11d8b6393fb9a59c305cae3db761fc89b089aadb81bedc23823a2a
ai-securityandroid-17baseband-securitycertificate-transparencycookie-theftdbscdevice-bound-session-credentialsdns-parsergeminigoogle-workspaceindirect-prompt-injectioninfostealeripilummac2memory-safetymerkle-tree-certificatespixelpost-quantum-cryptographypqcquantum-safe-httpsrustvrpvulnerability-rewards-program

What happened

Collection of Google Security Blog posts (Mar–Apr 2026) covering active AI and platform threats and mitigations. Key points: Google Threat Intelligence finds Indirect Prompt Injection (IPI) a prioritized, real-world attack vector and reports a web-wide sweep to detect IPI patterns; Google Workspace/Gemini defenses are being continuously improved for IPI resistance. Chrome/Account teams announced Device Bound Session Credentials (DBSC) rolling out to Chrome 146 on Windows (macOS soon)—a proactive mitigation that makes exfiltrated cookies unusable and defends against infostealers (e.g., LummaC2)

Why it matters

A reviewed impact interpretation has not been published for this record.

Evidence and limitations

Source ID
google_security_blog
Record identifier
3189ee2abb11d8b6393fb9a59c305cae3db761fc89b089aadb81bedc23823a2a
Enrichment time
2026-05-10T13:23:37Z
AI-assisted enrichment
Yes

This record may overlap with other records. Its enrichment can be incomplete or wrong, and machine assistance was used. Validate consequential decisions against the linked source and your own environment.