Helen Toner Discusses U.S.-China AI Race at Aspen Ideas Festival
CSET Georgetown Emily Tavenner
Helen Toner talked US-China AI competition at Aspen Ideas Festival in June. The panel pushed back on the simplistic 'race' framing everyone uses.
Based on reporting by CSET Georgetown, Emily Tavenner — read the original for the full story.
Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error
Helen Toner, who runs Georgetown's Center for Security and Emerging Technology, sat down at the 2026 Aspen Ideas Festival for a panel with a title that basically writes its own headline: "The AI Race: If We Don't, China Will." She was joined by Alvin Wang Graylin, a global technology strategist, and Melanie Hart from the Atlantic Council's Global China Hub, with Aspen Digital's Vivian Schiller moderating.
The conversation didn't just rehash the usual chip-export-controls talking points. Instead, the four dug into where the U.S. and Chinese approaches to AI actually diverge — not just in compute or capital, but in how each country builds, deploys, and governs the technology. That distinction matters more than most coverage of this topic admits. Framing everything as a two-horse race tends to flatten real differences in strategy, risk tolerance, and institutional strengths into a single scoreboard.
Toner has spent years arguing against oversimplified narratives in AI policy, and this panel fits that pattern. Rather than treating Chinese AI progress as an undifferentiated threat, the discussion apparently probed specific weaknesses on both sides — areas where Beijing's centralized approach might backfire, and where America's more fragmented, market-driven system creates its own vulnerabilities.
CSET only released this as a short recap pointing to the full video, so the granular arguments aren't spelled out in detail here. But the panel's existence, at a marquee venue like Aspen, signals that the framing fight — race versus something more nuanced — is still very much unresolved among the people shaping U.S. policy.
My take — AI-written commentary, not fact-checked reporting
I'll say what I always say: "race" framing is great for headlines and terrible for policy, because it pushes everyone toward matching capability numbers instead of asking which risks actually matter. Toner's been one of the few voices in D.C. willing to say that out loud, and I'd rather have ten more panels like this than another chip-export press release.
Read more about this at: CSET Georgetown
