Episode summary
🤠 LIVE from TX with TK, Jon Gray & Scott Wu
Summary
The conversation centered on how AI is reshaping capital spending, infrastructure, and corporate workflows. A guest from Blackstone argued that the current AI buildout is closer to the late 19th-century railroad era than to the telecom bubble because demand is already running ahead of supply, long-term contracts are backing projects, and the bottlenecks are chips, power, entitlements, and capital. He said Blackstone is seeing model spending rise 20-fold in a year and expects the next wave to extend from compute into energy, defense, pharmaceuticals, manufacturing, and robotics.
A guest from Micron made the case that memory has become a strategic AI input, not a commodity, and said the company is expanding aggressively in the United States with more than $250 billion of investment across Boise, Idaho, Clay, New York, and Virginia. He described AI as improving internal engineering speed, helping designs move twice as fast, while customer co-design is becoming more important as roadmap cycles stretch several years.
Later interviews shifted to how AI changes wealth management and software agents. A guest from Morgan Stanley argued that private and public markets should be seen as a continuum of disclosure and sophistication, and said the firm is rolling out AI tools first for advisers, then for clients. A guest from Cognition said Devon Bots are persistent agents that can own outcomes, not just tasks, and described security, routing, and post-training as key advantages.
Key takeaways
- Blackstone’s John Gray said AI infrastructure looks more like a contracted, supply-constrained industrial buildout than a speculative bubble, with model spending up 20-fold in a year.
- Gray argued that compute can be redeployed across AI use cases, while power, chips, and entitlements are the real constraints on the next phase.
- Micron’s Scott said the company is putting more than $250 billion into U.S. manufacturing, including Boise, Clay, and Virginia, because customer proximity and supply security matter more now.
- Scott said AI is already speeding Micron’s engineering work, cutting design time in half and enabling twice as many designs with the same headcount.
- Morgan Stanley’s Jed Finn said the firm is moving AI into wealth management in phases, starting with advisers and expanding to clients through curated agent workflows.
- Cognition’s Scott Wu said Devon Bots are built to own outcomes, and the company is focusing on post-training, product polish, and security rather than full pre-training.
Notable quotes
“trillions of dollars are needed to build. Yeah. And that changes things. And so”
“our companies, their spending on models is up 20fold in the last year. Okay”
“we're majorly focused in expansion across the globe certainly especially in the US with more than 250 billion of investment coming”
“these persistent agents that can actually own outcomes instead of tasks”
Watch the episode
Watch “🤠 LIVE from TX with TK, Jon Gray & Scott Wu” on YouTube
Summary, takeaways and quote selection by Octagon. Quotes are short excerpts; the full conversation is in the episode.
More from Tbpn
-
Goldman Sachs CEO Reacts to Jensen Huang's Potential Return to NVIDIA -
Will Jensen Return? FAA Meltdown, Amazon Casts Pattinson as Bond | Diet TBPN -
Big Week, Ground Stop, Not Just Another Technology, Robert Pattinson is James Bond -
NVIDIA Won’t Be the Same Without Jensen Huang — Tae Kim -
Jensen Huang Resigns as NVIDIA CEO | Diet TBPN -
TSMC Hacked, Vibes Down Outlook Down, 41 Days Till Christmas, Nvidia v2
Tbpn, this episode and its artwork belong to John Coogan & Jordi Hays. Octagon is not affiliated with the show; this page is an independent summary.