Publications: Hodan Omaar

U.S. AI policy prioritizes training compute while overlooking inference—the compute needed to deploy models effectively. As AI progress shifts toward optimizing inference, policymakers must adapt by supporting global deployment, refining export controls, and promoting energy-efficient AI to maintain U.S. competitiveness.

Virginia Governor Glenn Youngkin will decide on HB 2094, a flawed AI bill with inconsistent rules and unworkable enforcement. Signing it would add to a costly patchwork of state laws without improving outcomes.

Decades of investment and world-class research have put the U.S. at the forefront, particularly in quantum computing, but a fragmented approach and limited focus on sensing and communication create strategic gaps. Collaboration with the ROK could help strengthen supply chains, accelerate commercialization, and bolster U.S. leadership in an increasingly competitive global landscape.

The California Privacy Protection Agency (CPPA) has proposed regulations that would impose new requirements on businesses using automated decision-making technology.

The UK’s copyright system has long prioritized economic interests over creators’ control, allowing industries to learn from creative works without permission. Critics now opposing AI training have benefited from these same principles. A fair debate should acknowledge this precedent rather than selectively restricting AI.

DeepSeek’s breakthrough is a wake-up call that China’s AI capabilities are advancing faster than Western conventional wisdom has acknowledged.

The bill’s heavy-handed approach risks creating more problems than it solves, prioritizing bureaucratic hurdles over meaningful progress in fairness and accountability.

The United States has seen fragmented efforts in AI regulation. Lacking coordination, these efforts failed to create a unified strategy, leaving the U.S. without a clear voice on AI.

This post is part of our ongoing series on China’s AI unicorns. Moonshot AI, a Beijing-based startup, rose quickly in China's AI market with its flagship chatbot Kimi, known for processing 2 million characters and a consumer-focused approach.

Five Chinese generative AI start-ups, known as "AI Tigers," have achieved unicorn status by 2025, highlighting China's rapid innovation and distinct strategies in the global AI race. This blog series delves into each of these five startups.