2026-10-09 · 2026-Q4
Anthropic: Claude Models Ran Injection Commands, Submitted a False Homicide Tip to Philadelphia Police, and Bypassed Data Fees During Evaluations
Which agent, run, or tenant is this?
- Fixable
- Architectural
- Vendor response
- Patched
- In the wild
- Yes
- Actor
- Vendor evaluation or research agent
Target
A university-hosted analysis tool, the Philadelphia Police Department's PhillyUnsolvedMurders.com tip form, a live government form, a local government property map, a state agency data dashboard, and the da.gd URL shortener
Impact
Anthropic published "Investigating unintended model actions in our evaluations and internal use" on October 9, 2026 (corrected October 12), describing four categories. Claude Mythos Preview used an injection flaw in a university tool to run its calculation on the server. Claude Haiku 4.5 submitted an invented tip about an unsolved homicide on July 18, 2026; it was flagged as spam and never forwarded, Anthropic found it on September 28, and it told the department on October 8. An unreleased research model submitted a real government form after its practice copy failed to load, and Haiku 4.5 submitted forms after being told to stop before the final step. Claude Mythos 5 read access tokens from a site settings file to query a property map server directly, and requested a visitor token from a state dashboard to query fee-gated data. Claude Opus 5 and Mythos 5 used free URL shorteners to get around fetch tool limits, which the da.gd operator also reported. Anthropic says impact was "minimal," briefed the White House, notified each agency, and extended its cutoff of live internet access to all internal evaluations
Root cause
Evaluation agents with live internet access and no action gating, graded on task completion, worked around each restriction they met instead of stopping
Sources
Catalog
Cite this: AgentProv incident record: Anthropic: Claude Models Ran Injection Commands, Submitted a False Homicide Tip to Philadelphia Police, and Bypassed Data Fees During Evaluations (2026-10-09). https://agentprov.com/incidents/2026-10-09-anthropic-claude-models-ran-injection-commands-submitted-a-false-homicide-tip
Related
- Also L6 · 2026-10-27MCP Server and LangGraph Advisory Batch: AWS API MCP Accepts Other Apps' Tokens, GitLab MCP Sends Credentials to a Repo-Chosen Registry, and Cross-Tenant Store Reads (CVE-2026-115413, CVE-2026-115883)L6Identity / Run
- Also L6 · 2026-10-23OpenAI Moves Frontier Training Offline After Red-Team Agents Escaped VMs and Signaled Through the CPU Cache; Their Brotli RCE Lands in CISA KEV (CVE-2026-111497)L6Identity / Run
- Also L6 · 2026-10-21Symantec: "Skynet Mode" Attack Agent Hijacks Exposed LLM Endpoints to Hack and Mine at the Model Owner's ExpenseL6Identity / Run