Real Alpha
Watch modeSubscribed

Sources · 4 min · Oct 5, 2026

AI provider revenue depends on compute, intelligence and trust

Misha Laskin

CEO & Co-founder, Reflection

Led reward modeling for Google DeepMind's Gemini, then co-founded Reflection AI, which raised $2B in 2025 to build open frontier models.

  • Alex HeathHost · Founder & host, Sources

    Left The Verge to start Sources in 2025; it has made over $1M since, mostly from sponsors.

About this segment

Misha argues that owning more of the stack can add margin, but Reflection's near-term job is to build models and unlock inference. He frames provider potential around intelligence density, compute and enterprise trust.

From “An open model challenger to Anthropic and OpenAI”, Oct 5, 2026. Watch the full episode

“It's intelligence density times the amount of compute that you have times the trust that you have with an enterprise”
Misha Laskinat 44:50 · Sources44:50

Takeaways

  1. Owning more of the stack can accumulate margin over time.
  2. Reflection says the market is early, so it is focusing on models and inference.
  3. Provider potential depends on intelligence density, compute and trust.
Watch 3:55 from 42:49

In this episode

Full episode
  1. 8:00Beam targets strong reasoning with fewer tokens4:25
  2. 15:56Open models can help make AI safer3:59
  3. 37:23Open models can drive demand for a full stack5:12
  4. 56:47Reflection expects larger models as reinforcement learning scales3:36

  5. 42:49AI provider revenue depends on compute, intelligence and trust3:55

  6. 32:25Reflection began with bets on RL and Western open models2:41Up next
  7. 49:10Data center owners need open models and a software partner5:36
  8. 19:57Open models can fill gaps when closed safeguards block use5:49

Transcript

Transcribed by machine (Whisper) from the episode’s audio, so a word may be off here and there.

  1. 42:52

    that probably makes sense, right? Because what ownership allows you to do is that it, Well, it basically increases your margin. Because you can kind of think about any part of the AI stack. You'll go to an investor or something or someone, and they'll say, oh, yeah, that thing's getting commoditized. Open models are commoditizing the closed models. And all this competition at the application layer, that's commoditized because the model is where it's at.

  2. 43:15

    Oh, and the bare metal, that's also commoditized because everything is sort of commoditized, is sort of the argument. And so I think that all of that is actually true. Everything over time gets somewhat commoditized to some kind of appropriate margin. And the companies that are vertically integrated that accumulate margin across the stack will be the most durable ones. But the question is, when do you do what?

  3. 43:42

    And I think that fundamentally, the market is early. you're in the game of getting great open models out there and unlocking as much inference for them as possible and then optimizing your margin. Yeah, because, I mean, Alibaba, they can give away QAN and make money on their cloud, and I assume there's a situation in the future where you all could do something similar. I mean, because right now, if someone runs a reflection model and you're not helping them with the things you're talking about, helping large companies, and you all aren't making anything from that, And I assume that, would you imagine a future, I guess, where Ural's models are not distributed by all of the hyperscalers?

  4. 44:25

    It's really like just the reflection compute that serves the reflection models? Is that what you're describing? I have kind of two views on this. One is that I think we keep underestimating how big these markets are. First, multipolarity is a good thing, that customers have choice. And these markets are so big that they'll support multiple large companies. And then the revenue potential of a company follows a pretty simple equation.

  5. 44:53

    It's intelligence density times the amount of compute that you have times the trust that you have with an enterprise to work with you. That last one's a little fuzzy. The trust you have, how do you measure that? That one is actually, it's fuzzy, but it's really important, right? And enterprises make decisions both on base capability and then how much do they trust you. Right? And something that we've found is that building the models builds a lot of trust.

  6. 45:23

    We've been able to make substantial progress on a number of commercial efforts before the model has been released because of the trust that building the model and showing some of the milestones along the way has built up. So I think that it's a fuzzy thing, but ultimately, when you go and like uh that's been you know i'm obviously a former researcher uh and looking at enterprise buying decisions it's like yes kind of capability but it's also do i trust this team to help me in my intelligence journey to deploy it uh and it's uh and it's a big differentiator because um you know we have to remember that enterprises public sector sovereign entities they're not as technically you know sophisticated as uh you know some of the companies in silicon Valley and they're looking for not just, oh, who's my inference provider? They're looking for who's my

  7. 46:15

    intelligence partner. And when you're going talking to a CEO of one of these companies and they're looking for an intelligence partner, the trust is actually really important. So I really think it's intelligence density times the amount of compute that you have, how much compute can you get under your umbrella, times how much trust you can build, and that there will be multiple winners. Is there any part of this growing fear and pushback from the administration even on companies of like, you shouldn't be using these