Episode summary
Bloomberg Surveillance TV: July 17th, 2026 | Bloomberg Surveillance
Bloomberg Surveillance
· · 25 min · Hosts: Tom Keene, Paul Sweeney, Lisa Mateo, Michael Barr · Guests: Aaron Kennon, Jeff Curie, Dan Ives
Summary
The discussion opened with a tech-led selloff and the new Moonshot Kimmy K3 model from China, which the panel treated as another possible DeepSeek-style shock. Aaron Kennon argued that the bigger question is not just model quality, but whether Chinese teams are creating more intelligence per unit of compute and whether that could pressure semiconductor demand, hyperscaler capex, and the return on invested capital across the AI buildout. He said the market’s AI trade is increasingly vulnerable to correlation risk, even as he noted that broader market breadth has improved and that banks and equal-weight indexes have been strong.
Jeff Curie shifted the focus to energy, arguing that oil prices are the noise while product prices are the real signal. He said bottlenecks across refining, shipping lanes, and inventories have created a shortage, not just a deficit, and that commodities are the best-performing asset class because the energy system remains underinvested and exposed to shocks from the Middle East and Russia. He also argued that China’s behavior matters because teapot refiners have strong incentives to return when margins are this high.
Dan Ives called the China model release a white-knuckle moment but said it would not change his core view that the U.S. remains ahead in tech. He argued that AI monetization will come less from models than from data, infrastructure, and the application layer, and said the industry is still early in the AI cycle.
Key takeaways
- Aaron Kennon said the China model news raises a real question about whether AI output per unit of compute is improving enough to pressure semiconductor demand.
- Kennon argued that AI and semiconductors remain tightly correlated, but broader market breadth has improved with equal-weight indexes and banks outperforming.
- Jeff Curie said crude is “noise” and product prices are the real signal, pointing to a severe shortage in global refined fuels.
- Curie argued that the energy complex is underinvested and vulnerable, with refinery outages, shipping risks, and low inventories amplifying shocks.
- Dan Ives said the AI spending cycle is still early and that most value will accrue in data, infrastructure, and applications rather than in the models themselves.
Notable quotes
“Leadership is steadily expanding beyond semiconductor manufacturers towards the broader AI ecosystem”
“it also questions you know how much capex have we spent this year across the hyperscalers what will be spent next year”
“I think it's avoidable if you if you don't just own the market, right”
“crude oil prices are the noise. Product prices are the signal”
“I think this is just a called a white knuckle moment. No different than a a mini deepseek moment to some extent”
How the episode opens
Bloomberg Audio Studios podcasts radio news. This is the Bloomberg Surveillance Podcast. I'm Jonathan Pharaoh along with Lisa Abramitz and Amarie Hordern.
Join us each day for insight from the best in markets, economics, and geopolitics. From our global headquarters in New York City, we are live on Bloomberg television weekday mornings from 6:00 to 9:00 a.m. Eastern.
Subscribe to the podcast on Apple, Spotify, or anywhere else you listen. And as always, on the Bloomberg terminal and the Bloomberg Business app, we begin this hour with stocks extending techled declines. Aaron Kennard of Clear Harbor Asset Management writing, "Leadership is steadily expanding beyond semiconductor manufacturers towards the broader AI ecosystem.
Aaron joins us now for more. Aaron, we've got to talk about the latest out of China. Moonshot's new Kimmy K3 model.
Maybe a lot of people aren't familiar with what's coming out of China. They're telling us as they might that this can rival the models coming out of anthropic and open AI. Erin, do you see this as another Deep Seek type moment or is this different?
Sure. Feels that way, Jonathan. Um, you know, clearly it's a question around, you know, how competitive is this Kimmy 3 model relative to uh the closed source models that we have here in the United States uh made by Enthropic, OpenAI, Grock and others.
Um it also questions you know how much capex have we spent this year across the hyperscalers what will be spent next year could we see the slope change um as perhaps the return on invested capital question comes into the four I mean the the other question that I I think is worth pondering too is Kimmy's made a model that's competitive with with US models and they do not have the leading uh chips that are uh designed here in the United States by the likes of Nvidia like the Blackwell chip or the Reuben chip.
From the episode’s captions: roughly the first two minutes, lightly cleaned. Keep watching on YouTube →
Topics
- global markets
- economy
- asset management
- investment strategy
- finance news
Companies and tickers mentioned
- Clear Harbor Asset Management
- Altis Partners
- Yorkville Ives & Co.
Watch the episode
Watch “Bloomberg Surveillance TV: July 17th, 2026 | Bloomberg Surveillance” on YouTube
Summary, takeaways and quote selection by Octagon. Quotes are short excerpts; the full conversation is in the episode.
More from Bloomberg Surveillance
-
Stocks Rise as Inflation Data Ease Fed-Hike Worry | Bloomberg Surveillance -
Bloomberg Surveillance TV: November 13th, 2026 | Bloomberg Surveillance -
Bloomberg Surveillance TV: November 12th, 2026 | Bloomberg Surveillance -
Jensen Huang Ousted by Nvidia Board as CEO | Bloomberg Surveillance -
Bloomberg Surveillance TV: November 11th, 2026 | Bloomberg Surveillance -
AI Stocks Rebound as Fed Hike Bets Return | Bloomberg Surveillance
Bloomberg Surveillance, this episode and its artwork belong to Bloomberg. Octagon is not affiliated with the show; this page is an independent summary.
