Can AI Chip Demand Outrun the Power Crunch?
For wealth professionals following AI infrastructure, it frames the bottleneck as people, power and politics rather than chip demand alone, and distinguishes delay from cancellation in data-center projects.
Summary
By the real estate & infrastructure deskMorgan Stanley specialists discuss the gap between AI chip demand and the U.S. power and data-center buildout, including how midterm outcomes in Texas and Ohio may delay projects. They describe the estimated gigawatt shortfall rising from 30 into the mid-30s by 2028, with much bigger deployment in 2029 acting as a relief valve, while the outlook for AI semiconductors remains bullish.
- The discussion centers on a disconnect between the AI power gigawatt shortfall and a bullish outlook on AI semiconductors.
- Morgan Stanley had estimated a 30-gigawatt shortfall between now and 2028 and revised it into the mid-30s after the midterms.
- Gubernatorial outcomes in Texas and Ohio are seen as delaying, not cancelling, data-center projects.
- The speakers say 2029 deployment is much bigger and acts as a relief valve.
Similar items
Same topicGet the weekly 3 to watch
Free, in the Morning BriefEvery week the Morning Brief picks three videos and podcasts worth your time, with why each one matters.
Free. Every morning. Unsubscribe from any edition in one click.
Videos and podcasts are embedded from their publishers. Selections and summaries are editorial and are not investment advice.




