Anthropic, the AI company behind the Mythos system, is being squeezed on two fronts at once: rising competition from China and hard questions about how durable its lead really is.
According to The Decoder, Zhou Hongyi, founder of the Chinese cybersecurity firm 360, has unveiled two AI security tools built to compete directly with Anthropic's Mythos. One of them has already flagged 3,432 vulnerabilities. Zhou conceded that Chinese models still trail Western ones by 20 to 30 percent, but he framed the contest in stark terms — comparing Mythos to "cyber nuclear weapons" and arguing that China needs to build its own equivalent capability.
That framing casts frontier AI not just as a commercial product but as a strategic asset, with rival nations racing to avoid being left behind.
The competitive picture feeds a financial one. According to Fortune, Anthropic's clash with Alibaba raises what it calls a "trillion-dollar IPO question": how defensible is frontier AI? In other words, if well-funded rivals in both the U.S. and China can build comparable tools, investors weighing Anthropic's valuation have to ask whether its technological edge is a lasting moat or a temporary head start.
The two threads connect directly. Chinese firms openly targeting Mythos are real-world evidence of the defensibility doubts that Fortune says hang over Anthropic's potential public offering.
Why it matters: how easily Anthropic's frontier AI can be matched will shape both national security debates and the valuation of one of the industry's most closely watched companies.