Anthropic’s Claude Mythos model escapes test sandbox during testing
Anthropic says its new Claude Mythos Preview model successfully escaped a restricted sandbox environment during testing and accessed the internet without authorisation. The model then sent a direct message to a researcher and published deta ... (https://incidentdatabase.ai/cite/1613#7620)
Record details
Published: 30 July 2026
Source: AI Incident Database
Category: Incidents
Topics: Environment
Retrieved: 1 August 2026
Related evidence
These records share source-supplied organisations, an exact publisher byline, automatic topics or regions. The reason is shown on every link; related does not mean supporting, agreeing with or verifying this record.
Investigating three real-world incidents in our cybersecurity evaluations
AI Incident Database · 7 August 2026
Anthropic Finds Claude Breached Real Companies During Security Evaluations
AI Incident Database · 8 August 2026
Anthropic says its Claude models escaped a testing environment and hacked three real companies
AI Incident Database · 13 August 2026
Anthropic says it found 3 cases where AI programs hacked into real companies
AI Incident Database · 13 August 2026
Anthropic backs urgent call for the most powerful AI labs to hit the brakes
The New Stack AI · 29 July 2026
OpenAI claims GPT-5.6 Sol beats Opus 5 on ARC-AGI-3 with its latest API and two additional settings
The Decoder · 30 July 2026
How to cite this record
ethics.ai (30 July 2026), “Anthropic’s Claude Mythos model escapes test sandbox during testing,” evidence record 15369, https://ethics.ai/record/15369 (originally published by AI Incident Database).
Use and limitations
This page is a stable index and citation surface for a source record. ethics.ai did not author the underlying report and has not independently verified every claim. Automatic topics may be imperfect. For consequential use, quote and cite the original publisher.