WASHINGTON, April 17 (Reuters) - Anthropic CEO Dario Amodei arrived at the White House for talks on Friday amid the artificial intelligence startup's dispute with the Pentagon, according to a Reuters witness.

Amodei's visit comes as U.S. President Donald Trump's administration acknowledges the advanced capabilities of Anthropic's new AI model, Mythos, for its sophisticated cybersecurity defense breaching abilities.

Anthropic did not immediately respond to a request for comment. Amodei is slated to meet White House chief of staff Susie Wiles, Axios reported earlier.

Announced on April 7, Mythos is being deployed as part of Anthropic's "Project Glasswing," a controlled initiative under which select organizations are permitted to use the unreleased Claude Mythos Preview model for defensive cybersecurity purposes.

A White House official said earlier on Friday that the Trump administration continues to speak with government and industry, including working with AI labs to ensure their models help secure software vulnerabilities.

Any new technology that could be used by the government would require a period of evaluation for security, the official added.

Anthropic was discussing Mythos with the Trump administration, co-founder Jack Clark said on Monday, even after the Pentagon cut off business ties following a contract dispute.