AI threats in the wild: The current state of prompt injections on the web
2026-07-24T07:23:44Z•667c8af02d53fcccfad1d3e3be1c280f922654220efe723a0a81653bc9907734
AI threatsAndroid 17ChromeDBSCDNS parserGeminiGoogle WorkspaceIPILLM securityLummaC2Merkle Tree Certificates (MTC)PQCPixel basebandRustVRP 2025baseband securitycookie theftdevice bound session credentialsindirect prompt injectioninfostealermemory safetypost-quantum cryptographyprompt injectionquantum-safe HTTPSvulnerability rewards program
What happened
This collection of Google Security Blog posts covers several high-level threats and defensive initiatives rather than specific disclosed vulnerabilities. Key points: Google Threat Intel flagged Indirect Prompt Injection (IPI) as a top AI-era attack vector, performed broad web monitoring for IPI patterns and emphasizes continuous, layered mitigations (notably in Google Workspace/Gemini). Chrome and Google Account teams launched Device Bound Session Credentials (DBSC) into public availability on Windows (Chrome 146) with macOS coming, to proactively prevent session-cookie abuse by infostealer/ma
Why it matters
A reviewed impact interpretation has not been published for this record.
Evidence and limitations
- Source ID
- google_security_blog
- Record identifier
- 667c8af02d53fcccfad1d3e3be1c280f922654220efe723a0a81653bc9907734
- Enrichment time
- 2026-07-24T07:23:44Z
- 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.