The answer lives in this podcast
How far ahead is the United States from China in AI development, and is that lead secure?
The United States currently holds a lead of approximately six to eight months over China in AI development, an improvement from the previous four to six month gap. However, Katrina Mulligan emphasizes that this advantage is far from guaranteed—China possesses talented engineers and fierce determination to win, and the U.S. must treat this competition with the same intensity that China does.
The gap between American and Chinese AI capabilities has widened, but the margin remains measured in months, not years. Mulligan discusses this dynamic in depth during the episode, stressing that the speed at which both nations are advancing makes complacency a critical vulnerability.
What makes this lead precarious is not merely the capabilities gap, but the velocity of competition. At OpenAI, as Mulligan explains, a single month of development often feels equivalent to a quarter of progress at traditional organizations. This acceleration means the U.S. advantage could compress or even reverse if America loses focus or investment momentum.
The risk of overconfidence in a sprint mentality
Mulligan's central concern is not just about the raw technical lead, but about the mindset driving sustained competition. China's approach treats AI development as a matter of national strategy, with government and industry aligned toward a singular objective. The American system, by contrast, often treats technological advancement as a business problem rather than an existential competition.
She argues that the U.S. cannot afford to view this as a secured victory. Talented Chinese engineers and a centralized policy framework create conditions for rapid closure of any gap. The episode explores what this dynamic means for American AI leadership and the policy changes needed to maintain it.
"I honestly think that I am at least 30% more effective, maybe more than I was a year ago, because of how I've matured my use of these tools."
Katrina Mulligan — Head of National Security Partnerships at OpenAI. Previously a senior defense and national security official at the Pentagon, National Security Council, and Department of Justice, where she oversaw special operations at the Pentagon. She transitioned to OpenAI after describing the move as the hardest professional transition of her career, moving from one of the most hierarchical organizations on Earth to its opposite. Her dual perspective on both Washington's national security apparatus and frontier AI development makes her uniquely positioned to assess AI's geopolitical implications.
Key takeaways
- The U.S. AI lead over China has expanded to 6–8 months, but remains vulnerable and contingent on sustained commitment.
- China's centralized approach to AI development, with direct government involvement, creates structural advantages in closing capability gaps rapidly.
- Speed of innovation at frontier AI companies means traditional measurement of advantage must be recalibrated—months now represent what would take quarters elsewhere.
- Winning the AI race requires the U.S. government and industry to adopt the same intensity and strategic focus that China has already deployed.
See also
How does China's approach to AI differ from the United States, and could America build the best AI and still lose?
Mulligan identifies three key differences: China has much higher public trust in AI, the Chinese government and CCP have a more direct seat at the frontier, and the competition between the two nations mirrors earlier races in ways that demand American seriousness.
What is a 'model garden' and why does Katrina Mulligan consider it a mistake in AI adoption strategy?
A model garden is the practice of making 12 to 20 different AI models available in a single place so employees can choose among them. Mulligan argues this approach diffuses focus and slows organizational transformation compared to a standardized, single-model strategy.
What is the single biggest predictor of a successful AI transformation in an organization according to OpenAI's implementation experience?
Mulligan says the biggest predictor is the extent to which the C-suite personally uses AI in their own work. A bottom-up approach without executive adoption consistently fails to drive organizational change.