The issue between Anthropic and the Ministry of War is having a significant impact on Google and OpenAI. *** The AI industry is seeing a major shift in the recruitment landscape over the "ethics" of military use. A new battle has begun. Within Google and OpenAI, signatures for the petition "We Will Not Be Divided," opposing the military use of AI (mass surveillance, inhumane autonomous weapons), are rapidly increasing. (As of 2/28: Google 578 + OpenAI 93 = over 671 total) notdivided.org What is particularly noteworthy is not the number of signatories, but the "quality of the signatories." ■ Signatories who are at the "core of intelligence" • Google side • Tom Schaul: A world authority on reinforcement learning. Lead author of the DQN improvement. • Rif A. Saurous: Senior researcher supporting the scaling foundation of Gemini. • OpenAI side • Karl Cobbe: Contributor to the "thought process reward" of the latest inference model (o1). • Leo Gao: An expert in "interpretability" who exposes the black box of AI. This is not the idealism of a young person. It is existential resistance from the core group designing the "brain" of AI. ■ On the other hand, there were zero signatures from xAI. This contrast is brutally clear. xAI advocates "zero bureaucracy" and "wartime speed," and is comprised of people who accept military use as "lawful use." Highly capable individuals, fed up with the "abundance of safety meetings" at Google/OpenAI, are shedding their ethical brakes and moving towards "the fastest implementation." ■ In 2026, the talent market will "polarize." This ethical conflict is directly triggering a "mass migration of AI talent." 1. [Ethical stronghold: Anthropic] Rejecting government pressure, even at the risk of being designated a "supply chain risk." A "safety-first group," like the Google/OpenAI signatories, is gathering in search of conscience. 2. [The National Spear: xAI / National Defense AI] An accelerationist group that believes "technology is neutral; winning is justice" has gathered. Fueled by military funds, they are using overwhelming computing resources to dramatically advance intelligence. 3. [The Giant Caught in the Middle: Google / OpenAI] Drifting between national security pressures and employee resistance. The top management is defecting to both extremes in search of "ideological alignment," facing the risk of organizational hollowing out. ■ Conclusion AI is no longer simply a technological competition. The clash of "national ideology × corporate culture × ethics" has become the filter that determines which talent is allocated to which company. We are now at a critical crossroads where AI will either become a "tool of peace" or a "weapon of victory."