AI threats in the wild: The current state of prompt injections on the web
2026-08-07T01:23:33Z•3b5e5ff48b7dce8a284225087f69c50a798fccbaee66bc003c3d1af09f53c946
AI agentsAI securityAndroidCertificate TransparencyChromeDNS parserDevice Bound Session CredentialsGeminiGoogle WorkspaceHTTPSLLM securityLummaC2Pixel modemRustcookie theftindirect prompt injectioninfostealermemory safetypost-quantum cryptographyremote code executionsession theftvulnerability rewards program
What happened
Google reports on current security efforts covering real-world indirect prompt injection against AI agents and Workspace Gemini, browser session-cookie theft protections through Device Bound Session Credentials, memory-safe Rust adoption in Pixel modem firmware, post-quantum cryptography migration for Android and HTTPS, and findings from its 2025 Vulnerability Rewards Program. The most immediate threat focus is indirect prompt injection and credential theft, while the cryptography and Rust initiatives are proactive hardening measures.
Why it matters
A reviewed impact interpretation has not been published for this record.
Evidence and limitations
- Source ID
- google_security_blog
- Record identifier
- 3b5e5ff48b7dce8a284225087f69c50a798fccbaee66bc003c3d1af09f53c946
- Enrichment time
- 2026-08-07T01:23:33Z
- 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.