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

2026-06-28T01:23:49Zd83dd8d003486b9a447f84bf1f9668813cc4a13803c310377f178d5efb1ece54
AI agentsAndroid 17Certificate Transparency (CT) working group (PLANTS) notations?DBSCDNS parserGeminiGoogle WorkspaceIPILLM securityLummaC2MTCMerkle Tree CertificatesPLANTSPQCPixel basebandRustcookie theftdevice bound session credentialsindirect prompt injectioninfostealermemory safetymodem securitypost-quantum cryptographyprompt injectionquantum-safe HTTPS

What happened

This collection of Google Security Blog posts (Mar–Apr 2026) highlights several active and planned security initiatives: Google Threat Intelligence found real-world instances of Indirect Prompt Injection (IPI) on the public web and describes patterns and mitigations; Google Workspace and Gemini receive continuous defenses against IPI and agentic attacks. Chrome is rolling out Device Bound Session Credentials (DBSC) to prevent stolen/exfiltrated cookies from being reused (Windows available in Chrome 146, macOS coming), countering advanced infostealers such as LummaC2. Pixel firmware security is

Why it matters

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

Evidence and limitations

Source ID
google_security_blog
Record identifier
d83dd8d003486b9a447f84bf1f9668813cc4a13803c310377f178d5efb1ece54
Enrichment time
2026-06-28T01:23:49Z
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.