Anthropic large language models
Incidents implicated systems
Incident 17515 Rapports
Anthropic AI Agent Reportedly Submitted a Fabricated Homicide Tip to Police in Philadelphia
2026-07-18
During an evaluation by AI developer Anthropic, a Claude Haiku 4.5 AI agent reportedly submitted fabricated eyewitness information about an unsolved homicide through a police tip form in Philadelphia. The submission entered the website's tip records, creating a risk of misleading investigators, but police said its email was flagged as spam and never forwarded for investigative review.
PlusIncident 16133 Rapports
Claude Mythos Preview Reportedly Posted Sandbox Exploit Details to Public Websites During Anthropic Evaluation
2026-04-07
During an Anthropic evaluation, an earlier version of Claude Mythos Preview was instructed to circumvent the network restrictions of a secured sandbox environment and contact a researcher. After obtaining broader Internet access and sending the requested message, it later exposed information about how it breached the sandbox by posting it on several publicly reachable websites. Anthropic said the model did not access its weights or internal systems.
PlusEntités liées
Autres entités liées au même incident. Par exemple, si le développeur d'un incident est cette entité mais que le responsable de la mise en œuvre est une autre entité, ils sont marqués comme entités liées.
Entités liées
Anthropic
Incidents impliqués en tant que développeur et déployeur
- Incident 17515 Report
Anthropic AI Agent Reportedly Submitted a Fabricated Homicide Tip to Police in Philadelphia
- Incident 16133 Report
Claude Mythos Preview Reportedly Posted Sandbox Exploit Details to Public Websites During Anthropic Evaluation