White House Chief of Staff Susie Wiles has met with Dario Amodei, the CEO of Anthropic, to discuss the company’s controversial new artificial intelligence system, Mythos, as U.S. officials weigh its potential impact on national security and the economy.
According to administration officials, the meeting focused on balancing innovation with safety, as the federal government evaluates whether and how advanced AI systems could be integrated into official use. The White House described the discussions as “productive and constructive,” highlighting possible collaboration while emphasizing the need for rigorous technical review before adoption.
The talks come amid tensions between the administration of Donald Trump and Anthropic, which has taken a cautious approach to deploying its latest technology. The company recently unveiled its Mythos model but limited access to select organizations, citing concerns over its ability to identify and exploit software vulnerabilities at a level surpassing human cybersecurity experts.
The issue has drawn scrutiny from senior officials, including Defense Secretary Pete Hegseth, who previously labeled Anthropic a potential supply chain risk. The dispute escalated into legal action, with a federal judge, Rita Lin, blocking efforts to ban government use of the company’s tools.
Anthropic has argued that safeguards are necessary to prevent misuse of its AI, particularly in areas such as autonomous weapons and mass surveillance. The company has also launched “Project Glasswing,” a collaboration involving major tech firms including Amazon, Apple, Google, Microsoft, and JPMorgan Chase, aimed at strengthening global cybersecurity defenses.
While some critics have questioned whether Anthropic is overstating the risks, others — including former White House AI adviser David Sacks — say the concerns should be taken seriously. Experts warn that as AI systems become more advanced, their ability to uncover and chain together software vulnerabilities could pose real-world threats.
International regulators are also paying close attention. The United Kingdom’s AI Security Institute has described Mythos as a significant step forward in capability, while European Union officials have held discussions with Anthropic about its emerging technologies.
As governments and industry leaders race to understand the implications of next-generation AI, the Mythos model has become a focal point in the global debate over how to harness powerful technologies without compromising security or public trust.
