Advanced system moves from private testing to wider access
Anthropic has made a powerful version of its AI system available to selected public users. The model, called Claude Mythos 5, was previously restricted due to concerns about its strength and potential misuse.
The company says this version includes controls and limits, but also admits it is more capable than any earlier release.
Some experts in technology, finance, and government circles had already warned that the system could create risks if used without strict oversight.
Why the model raised concerns
Early private testing of Claude Mythos showed strong performance in technical tasks. Some testers reported that it could identify major weaknesses in digital systems.
In response, the company limited early access to a small group of organizations. These groups focused on cybersecurity and infrastructure protection.
Reports suggest that thousands of security issues were discovered during testing, which increased both interest and concern around the system.
New safeguards and controlled access
The updated version, Claude Mythos 5, is now being offered to more users under controlled conditions.
The company states that access is still limited to trusted organizations. Some safety restrictions are adjusted depending on how the model is used.
The system can also run tasks for long periods with minimal human input. This level of autonomy is one reason experts continue to discuss possible risks.
Industry reaction and wider debate
The release has intensified debate about how fast advanced AI should move into public use.
Some experts believe the technology can strengthen cybersecurity and help fix system weaknesses faster than humans alone.
Others argue that the pace of development is too fast and that stronger limits are needed before wider release.
A senior figure at the company recently said the industry has strong acceleration but lacks effective control tools to slow progress when needed.
