Episode summary
The Real Tech Crash Hasn't Even Started: Here's The Trigger | John-Clark Levin
The David Lin Report
· · 33 min · Hosts: David Lin · Guests: John-Clark Levin
Summary
John-Clark Levin argued that the recent tech selloff looks more like a temporary pullback than the deeper AI correction he expects later. He said the real problem is a growing gap between AI’s demo capabilities and its reliability in economically useful work, which forces companies into expensive “centaur” workflows where humans still supervise the models. In his view, markets will stay uneasy until AI can be deployed more autonomously.
He also downplayed the idea that the midterms will materially slow U.S. data center buildout, saying political noise may change where projects are located but not the underlying demand. On AGI, he said GPT-6 Astra was the first model to move his timeline meaningfully earlier, pulling his modal AGI year to 2028, while 2029 remains the point when AGI odds cross 50%. He added that continual learning is the biggest R&D problem now.
Levin said frontier inference is already profitable on a per-token basis, but the massive capex race is justified because firms are aiming at AGI’s long-term prize. He warned that if markets turn, Google and Meta could be in a stronger position than OpenAI and Anthropic, and he sees rising government and nationalization risk if AI becomes too powerful or too economically dominant.
Key takeaways
- Levin expects the next AI market correction to be “deeper and sharper” than the recent Nasdaq pullback, which he views as a temporary pullback for now.
- He said AI adoption is bottlenecked by reliability, not peak demo performance, so investors may stay disappointed until models can work autonomously.
- He thinks the midterms will not materially slow U.S. data center demand, though builders may face more local offset requirements.
- Levin said GPT-6 Astra pulled his AGI timeline earlier, making 2028 the modal year and 2029 the year when AGI odds cross 50%.
- He warned that Google and Meta could benefit from an AI correction because their non-AI businesses can fund capex while weaker rivals may struggle.
- He sees nationalization risk as real if one frontier lab approaches an economically dominant position or if a major safety incident occurs.
Notable quotes
“I expect the market correction when it happens will be significantly steeper than what we've seen over the past couple of weeks”
“Market speculation and hype are driven by peak capabilities, the best things AI can do in a demo”
“I do not think the midterms will have a very big effect on that”
“I still think 2029 is the year at which the chances of AGI cross 50%, but 2028 is probably the modal year now”
“The only way to do this seriously is to air gap the training clusters on which initial AGI pointing research is happening until models have gone through extensive safety testing”
How the episode opens
I expect the market correction when it happens will be significantly steeper than what we've seen over the past couple of weeks. I expect that to be deeper and sharper and more panic vibes, whereas right now it it feels like more of a temporary pullback. I was Sundar Pichai right now or Mark Zuckerberg right now, I'd be low key hoping for a severe AI market correction.
The Nasdaq is now down about 12% from its October peak to today, November 5th, when this interview was recorded. The sell off intensified earlier in the week right after the midterm elections and according to many analysts, it's now approaching bear market territory. Our next guest warned of a tech market correction before artificial general intelligence or AGI is officially here.
So is this the correction he's been warning us about and will it get worse? John Clark Levin joins us now. He's the head of research at Kurzweil Technologies.
John Clark is going to help us analyze the current market volatility, whether the AI data center build out is going to slow down, and the largest AI models that are pushing us one step closer to AGI. We're also going to examine the possibility of the government pausing AI growth and limiting the expansion of data centers, which has become a political rallying point lately. Right now on Kalshi, traders are predicting a 75% chance that four or more states will adopt data center moratoriums this year.
If you agree with those odds, a $50 trade could yield a payout of $65 if you're correct. This video is sponsored by Kalshi, the largest prediction market in the United States. Unlike a sportsbook, you're trading peer-to-peer on real world events from economic data to political outcomes and the price moves based on public opinion, not a house.
From the episode’s captions: roughly the first two minutes, lightly cleaned. Keep watching on YouTube →
Topics
- ai market correction
- tech stock selloff
- agi race
- data center spending
- ai safety and government intervention
Companies and tickers mentioned
- Kurzweil Technologies
Watch the episode
Watch “The Real Tech Crash Hasn't Even Started: Here's The Trigger | John-Clark Levin” on YouTube
Summary, takeaways and quote selection by Octagon. Quotes are short excerpts; the full conversation is in the episode.
More from The David Lin Report
-
Jensen Huang RESIGNS After Second Affair: Nvidia's $855B Wipeout, What Next? -
New Global Arms Race Begins: Who Wins Trillion-Dollar Defense Boom? | Juno Industries -
Collapse Of Tech? Rosenberg &. Bianco Debate AI Bubble, Soaring Bond Yields -
Buy The Dip Or Short? Trader Reveals Key Signals | Jason Shapiro -
Nvidia Spy Scandal: $450B Crash Just The Beginning? -
Brutal 2027: How The Next Financial Crisis Starts
The David Lin Report, this episode and its artwork belong to The David Lin Report. Octagon is not affiliated with the show; this page is an independent summary.
