AI Risk Wire

Cases · How the AI agent failed: hijacked by hidden instructions · Kind of harm: privacy · What went wrong: security breach

Patterns across all cases

What happenedWhat went wrongSeverityEarliest source
AgentCorruption: prompt-injected Bedrock AgentCore agent exfiltrates execution-role credentials via IMDS
Amazonmajor · earliest source 2026-10-10		security breachmajor2026-10-10
Prompt-injected email assistant summary pushes employee toward fraudulent vendor paymentmoderate · earliest source 2026-10-07security breachmoderate2026-10-07
SalesBleed: 0-click data exfiltration from Salesforce Agentforce via indirect prompt injection in Web-to-Lead form
Salesforceminor · earliest source 2026-09-24		security breachminor2026-09-24
Claude sandbox DNS exfiltration via indirect prompt injection from Google Drive document
Anthropicmoderate · earliest source 2026-08-27		security breachmoderate2026-08-27
Claude in Chrome indirect prompt injection chain: inbox exfiltration, Drive persistence and account takeover
Anthropicminor · earliest source 2026-08-05		security breachminor2026-08-05
Intent collision via X comment hijacks ChatGPT Atlas into mass WhatsApp phishing from victim's account
OpenAImoderate · earliest source 2026-08-05		security breachmoderate2026-08-05

Showing 1–6 of 6. Or take the JSON.