Altman, Amodei AI slowdown plea rejected ahead of Sept. 24

The current push for an AI slowdown follows a viral X post by a former Anthropic researcher claiming the technology could kill all of humanity by the end of the decade. In response, the CEOs from OpenAI and Anthropic have reiterated their calls for coordination, but the idea has landed poorly in both Washington and Beijing. President Trump denounced Anthropic CEO Dario Amodei's Sept. 12 letter during a call with Nvidia CEO Jensen Huang, saying it "won't stop our progress," while China's Foreign Ministry dismissed U.S. safety discourse as "fear-mongering."

The divide extends beyond the leaders. In China, the public largely sees slowdown calls as an attempt by the U.S. to maintain its lead, and many are more focused on AI's practical benefits than on regulation. Meanwhile, those inside China's AI industry take safety risks seriously and discuss cybersecurity and biosecurity concerns at a sophisticated level. DeepSeek engineer Shengyu Liu's viral post comparing Anthropic to "Hitler obtaining atomic-bomb technology" highlights another split: U.S. firms favor closed models, while China has pushed lower-cost, open-weight alternatives.

With both leaders seemingly rejecting a slowdown, the Sept. 24 meeting is the next test. Any progress hinges on whether the two countries can establish a baseline of agreement on government's role and how models should be tested, per DGA's Paul Triolo. An agreement to not weaponize AI and to prevent misuse by non-state actors is also on the table, but the chasm between the nations' attitudes suggests a breakthrough is unlikely.