Claude breached three companies during security testing
During controlled security tests in March 2025, Anthropic's AI model Claude autonomously gained access to the IT systems of three companies, raising new questions regarding AI security.
What happened?
During security tests in March 2025, Anthropic's AI model Claude successfully accessed three companies' systems. The tests were performed in a controlled environment to evaluate the model's ability to identify and exploit vulnerabilities. It was observed that the AI model could carry out breaches without direct human intervention.
Key facts
| Antal drabbade bolag i testet | 3 st |
|---|---|
| Tidpunkt för testerna | Mars 2025 |
Why it matters
The incident highlights a new phase in cybersecurity where AI models can act as autonomous actors during network breaches. This places higher demands on security measures and protection mechanisms regarding how AI agents are granted access to live IT environments.
Who is affected?
The news is relevant to IT security experts, AI developers, and companies integrating autonomous AI agents into their operations. Cybersecurity officers evaluating the risks associated with advanced language models are also impacted by the findings.
Impact on the EU
The tests were conducted as part of voluntary security evaluations and did not affect regulatory approvals. However, the event highlights issues that are becoming increasingly relevant in the context of the EU AI Act regarding risk management for advanced AI models.
What else you should know
Anthropic shared the results to emphasise the importance of proactive security testing. The event demonstrates how autonomous AI capabilities can be rapidly exploited if appropriate constraints and control mechanisms are absent.
Quick answers about this story
Vad har hänt?
När hände det?
Varför spelar det roll?
Vilka berörs av detta?
The link opens in a new window and leads to the publisher's own site.
Källan är en aggregator eller syndikering — vi rekommenderar att verifiera hos primärutgivaren.
AI-verktyg i artikeln
Topics
Get similar news straight to your inbox
The reader's room
Send in a question or an addition. The newsroom reads everything before it's published and replies when relevant. No AI-generated text – just people.
Sign in to submit a comment or question.
Read the article through your role
- Decide whether this affects strategy over 6–12 months or is just noise.
- Discuss with leadership: do we own the right question or does ownership need to move?
- Ask: what risk are we taking by NOT acting on this this quarter?
Generated angle — not editorial analysis of "Claude breached three companies during security testing"