AI’s Next Big Leap

April 28, 2026

AI Summary

5 min read

Tom Wigg, Head of Specialty Sales in the Americas at Morgan Stanley, interviews Stephen Byrd, Global Head of Thematic and Sustainability Research at Morgan Stanley, about the shift from linear to exponential improvements in AI frontier models. They discuss scaling laws driving these changes, their broad economic effects, compute constraints, and investor implications, emphasizing that markets have underappreciated the pace.

Scaling Laws and Nonlinear Progress

Byrd highlights a clear scaling law from recent years: increasing training compute by 10 times roughly doubles AI model capabilities. This relationship, evident in large language models (LLMs), predicts nonlinear leaps this spring and summer. Upcoming frontier models will handle a larger share of economic tasks with higher accuracy at very low cost. These advances will reveal underappreciated capabilities across industries, spreading disruption but also enabling new efficiencies. Byrd stresses that not all business models face the same fate—some will be disrupted, others supported or enhanced by AI, and a few may prove immune. Investors must evaluate each case thoughtfully rather than assuming blanket disruption.

Continue reading the full summary in the app — free to try.

Read Full Summary →

Free • No credit card required

Listen to Audio Summary Open in App

What you'll learn

1 (00:00) **Intro to Nonlinear AI Improvements** - Host Tom Wigg sets stage for exponential AI leaps beyond linear market assumptions
2 (00:31) **Frontier Models Obsession** - Byrd confirms nonlinear spring improvements in benchmarks playing out
3 (01:11) **AI Scaling Laws Defined** - Explains compute-to-capability relationship: 10x training compute yields 2x model performance
4 (01:50) **Industry Disruption Framework** - AI to enable/supersede tasks across economy at low cost, forcing investor scrutiny
5 (02:58) **Compute Constraints and ROI Debate** - Addresses hyperscaler CapEx spending vs. bearish stocks and potential cuts
6 (03:42) **Token Economics Model** - Hyperscalers/LLM devs see excellent returns even fully loading data center costs
7 (04:43) **Adopter ROI Math** - Enterprise LLMs replace 1.5 human hours ($55 savings) for ~$5/million tokens in agentic tasks

+ Full timestamped outline available in the app

Show Notes

Tom Wigg and Stephen Byrd discuss the accelerating pace of AI breakthroughs, the forces driving them and why the next phase of development may look very different from anything we’ve seen so far.

Read more insights from Morgan Stanley.

----- Transcript -----

Tom Wigg: Welcome to Thoughts on the Market. I’m Tom Wigg, Head of Specialty Sales in the Americas at Morgan Stanley, and a sector specialist in Technology, Media and Telecom.

We wake up every day to new AI product releases, so it’s easy to lose sight of the unprecedented non-linear improvement in AI capabilities. But things are about to get weird.

It’s Tuesday, April 28th at 8am in New York.

The market has been thinking about AI in linear terms. But we need to reframe that assumption of only incremental improvement and think about exponential improvement.

That was my takeaway from a conversation with Stephen Byrd, Global Head of Thematic and Sustainability Research at Morgan Stanley. In our conversation, we zeroed in on Stephen’s bull case for broader AI model improvements.

Tom Wigg: First, I want to talk about one obsession that you’ve been writing about for the last several months – is this idea that we’re going to see nonlinear improvements in the frontier models coming out this spring.

Stephen Byrd: Yes.

Tom Wigg: There’s been, you know, some big headlines around new models, benchmarks coming out publicly. Is this, you know, your bull case playing out on these models? And what are the implications?

Stephen Byrd: Yes! Absolutely, Tom. So we have, to your point, we are obsessed. And I know I’m not shy about that – with the nonlinear rate of AI improvement. It is the most important impact to so many stocks that I can think of in the sense that it can impact all industries, all business models. So, what we’ve been saying for some time is, if you look back over the last couple of years at the relationship between the amount of compute used to train these LLMs and the capabilities, we have a very clear scaling law.

And approximately the law is, if you increase the training compute by 10x, the capabilities of the models go up by 2x. Now, as you and I’ve talked about this a lot; just meditate on that for a moment. I think things are about to get weird in the sense that on the positive side, we’re going to see all kinds of underappreciated capabilities across many industries. So this disruption discussion, I think, is going to spread, but it’s also going to require investors to, kind of, be more thoughtful about what they do with that concept. Meaning you can’t sell everything. In the sense that AI will disrupt some businesses.

I actually think this is healthy in some ways because now it