AMD, Starcloud, Coatue..10 Hot Takes From The Biggest Names in AI

Summary

Sourcery with Molly O’Shea

Episode: AMD, Starcloud, Coatue...10 Hot Takes From The Biggest Names in AI

Date: July 19, 2026

---

Episode Overview

This episode of Sourcery captures the excitement and insight of the Raise Summit, gathering top investors, CEOs, founders, and technologists from across the AI universe. Host Molly O’Shea conducts rapid-fire interviews in a secret European location (spoiler: it’s probably somewhere near the borders of France and Germany), focusing on the “hottest takes” in AI. With guests from CO2, AMD, Starcloud, Black Forest Labs, General Intuition, SambaNova, Liquid AI, Factory, and Good Future Media, the episode navigates transformative trends in AI interfaces, infrastructure, open vs. closed models, device shifts, European tech’s emerging identity, and the evolving business of AI.

---

Key Discussion Points & Insights

1. The Death of the Keyboard & Mouse: The Rise of Agentic Natural Language Interfaces

• Guest: Max Cook (CO2)
• Timestamps: [01:04]–[05:20]
• Summary:
  • Max argues the mouse and keyboard are dying, soon to be replaced by natural language and voice-based interaction with agentic models.
  • Historic shifts in computing (mainframes → PCs → mobile) have always redefined the user interface; AI agents are next.
  • The average user now uses 11 apps per day, but with agentic ecosystems, this consolidates to just a handful, run primarily through APIs or direct interaction with the AI.
  • Quote:

    "We’ve always conformed ourselves to computers... now we’re finally entering the era where computers have to conform to us." — Max, paraphrasing Greg Brockman [02:40]

  • At CO2, investors already use voice prompting for improved AI-assisted workflows.
  • Predicts mainstream adoption will come as OpenAI and competitors move towards “super apps.”

2. Open Source vs. Closed Frontier: A False Binary?

• Guest: Max Cook (CO2)
• Timestamps: [06:35]–[11:03]
• Summary:
  • The real divide isn’t open vs. closed models, but different models for different needs.
  • Open source models win in production-scale, latency-sensitive workflows, while closed (frontier) models are still needed for cutting-edge R&D or new use cases.
  • Quote:

    "If you are trying to discover a new use case, you want the smartest model you can possibly get, you use the Frontier." — Max Cook [07:39]

  • The future may be hybrid: frontier models to “hone” workflows, open source for efficient production.
  • Raises provocative questions: What happens if OpenAI or Anthropic start selling their own older, fine-tuned, or cheap models as products?
  • Enterprise adoption of open source is growing rapidly—a recurring theme in later interviews.

3. Reframing AI Success: Data Quality > Model Complexity

• Guest: Pim (General Intuition)
• Timestamps: [13:21]–[14:20]
• Summary:
  • AI progress often just comes down to applying data science on better datasets.
  • Quote:

    "Most of AI is just glorified data science... Good models mostly are downstream from good data and simple things." — Pim [13:50]/[13:57]

  • Pim is bullish on truly great datasets and suggests industry fuss over models sometimes misses this point.

4. Is Yann LeCun Underrated?

• Guest: Pim (General Intuition)
• Timestamp: [14:21]–[14:27]
• Summary:
  • Pim jokingly supports the notion that Yann LeCun is "underrated" in the current AI hype cycle—"the friends just told me to say that."

5. Agentic AI Revolutionizes Workflows Faster Than Anyone Expected

• Guest: Mark, CTO of AMD
• Timestamps: [14:50]–[16:34]
• Summary:
  • Massive shift: CPUs and GPUs now used equally as AI moves to agentic workflows.
  • AMD is seeing huge acceleration in chip design using thousands of sub-agents—AI now drives daily operation.
  • Quote:

    "The ratio of CPU to GPU is becoming like 1 to 1... None of us anticipated how fast these agentic workflows would just take off." — Mark, CTO of AMD [15:29]

  • The summit is "bounded by the President of France and ends with Yann LeCun,” signifying the breadth of AI’s reach.

6. Open Innovation Is Critical for AI Progress & Safety

• Guest: Robin (Black Forest Labs)
• Timestamps: [17:13]–[18:16]
• Summary:
  • Open access (open source, open weights) is fundamentally good for the world and AI safety.
  • Fear-mongering about open models leads to closed models, which impedes progress.
  • Quote:

    "Open innovation is good for the world... accessible AI models are key." — Robin [17:26]

7. European Tech: Emerging, Not Yet Arrived

• Guest: Robin (Black Forest Labs)
• Timestamps: [18:41]–[19:13]
• Summary:
  • Europe’s tech scene is building, but scale and impact still lag behind the US.
  • Needs a mindset shift to push harder on innovation rather than just hosting “nice” events in castles.

8. Tokenising AI: Open Models’ Share of Compute is Surging

• Guest: Matan (Factory)
• Timestamps: [21:05]–[22:36]
• Summary:
  • Prediction: In the next 12 months, up to 90% of compute tokens (not necessarily cost or importance) will be spent on open AI models.
  • Quote:

    "At the beginning of the year... less than 1% of tokens were going to open models... In May it crossed the 10% threshold." — Matan [21:47]

  • By year’s end, enterprise usage of open models might hit 50%.
  • Open (especially non-US/non-Western—e.g., “Chinese”) models are cost-effective and increasingly performant.

9. SpaceX IPO & Orbital Data Centers Will Redefine Global Compute

• Guest: Philip Johnston (Starcloud)
• Timestamps: [24:06]–[25:51]
• Summary:
  • SpaceX will soon be recognized as the most undervalued IPO in history, poised to break $10T, monopolizing not just launches but enabling entire new industries—data centers in orbit, asteroid/lunar mining, comms.
  • Quote:

    "It’s like owning the railroads. Lots of businesses will be built on top of it..." — Philip Johnston [25:05]

  • Starcloud is moving to a massive facility for orbital compute (StarCloud 3), targeting 10MW per Starship launch.

10. Physical AI: Intelligence on Every Device, Cost-First Approach

• Guest: Ramin (Liquid AI)
• Timestamps: [32:03]–[38:39]
• Summary:
  • Liquid AI focuses on running foundation models at extremely low cost, even on resource-constrained devices (e.g., Raspberry Pi).
  • Three-Axis AI Future:
    1. Maximizing intelligence/capability
    2. Maximizing efficiency/lowering cost (not an afterthought anymore)
    3. Substrate—deploying AI anywhere (cars, phones, planes, not just data centers)
  • Quote:

    "Efficiency is not an afterthought; energy is not abundant... You have to invest and also supply chains are getting disrupted." — Ramin [34:55]

11. Clipping is the New Ad: The Human Touch in Content Virality

• Guest: Chris (Good Future Media)
• Timestamps: [27:02]–[30:54]
• Summary:
  • The business of clipping podcast/video content is central to reach and virality.
  • Key to virality: Grab attention in first three seconds, human psychology beats pure automation.
  • Quote:

    "The first three seconds need to grab the viewer in... If you don’t have [a hook], think, you know, restart, think of something else." — Chris [28:28]

  • Top-performing clips: millions of views for thought leaders on All-In Pod, 20VC, and others.

12. SambaNova: The Business Case for "Premium Inference" Chips

• Guest: Rodrigo (SambaNova)
• Timestamps: [38:45]–[41:02]
• Summary:
  • Announcing $1B Series F at $11B valuation, led by General Atlantic.
  • New SM50 RDU will deliver high-throughput, low-cost "premium inference" for billion- and trillion-parameter AI models.
  • Optimized inference means lower costs and better economics for cloud providers.
  • Quote:

    "When you’re running faster with the high accuracy models, you can charge more at a lower cost, allowing their service providers to generate more margins for their services." — Rodrigo [40:13]

---

Notable Quotes & Memorable Moments

• On the end of the mouse and keyboard:

  "You should throw them away. We’re going to... natural language interacting with agentic models."
  — Max Cook [02:59]

• On agents at AMD:

  "Every day I start with agents, they run my day, look at what's coming up... thousands of sub agents just doing incredibly complex chip design."
  — Mark, CTO of AMD [15:56]

• On Europe’s tech identity:

  "Europe is not known for having, I don't know, nice castles and stuff like that, but actually nice AI or nice technology."
  — Robin [18:41]

• On AI infrastructure evolution:

  "End state of AI inference would be... 10% open source... Or the combination of a frontier model for your hardest tasks and an open source for sort of production grade workflows. Like customer service..."
  — Max Cook [09:40]

• On data quality:

  "Good models mostly are downstream from good data and simple things."
  — Pim [13:57]

• On AI cost and deployment:

  "We are bringing foundation models to the physical world... so cheap that you can bring them on Raspberry Pis."
  — Ramin [32:21]

---

Quick Reference: Timeline & Segment Guide

• 00:56 Max Cook (CO2): Hottest take – keyboard/mouse are nearly dead, rise of agentic interfaces
• 06:35 Max Cook (CO2): Open vs. closed models, the production/innovation divide
• 13:21 Pim (General Intuition): Data quality, "glorified data science," and Yann LeCun shoutout
• 14:50 Mark (AMD): Agents revolutionizing workflows, CPUs and GPUs parity
• 17:13 Robin (Black Forest Labs): Open innovation and the state of European AI
• 21:05 Matan (Factory): Open model compute share skyrocketing
• 24:06 Philip (Star Cloud): SpaceX’s undervalued position and orbital compute future
• 27:02 Chris (Good Future Media): The business and psychology of social media clipping
• 32:03 Ramin (Liquid AI): Cheap, efficient AI on any device, the three-axis approach
• 38:45 Rodrigo (SambaNova): $1B round, premium inference chips for global AI scalability

---

Tone and Language

The episode vibrates with candid industry enthusiasm; Molly’s rapid, informal interview style brings out honest, sometimes provocative “hot takes” (with plenty of “Damn!” moments). The participants are unfiltered, speaking in a friendly, competitive, slightly irreverent manner that gives listeners front-row access to the cutting edge of AI innovation — alongside humor about European castles, heat waves, and the unglamorous logistics of AI summits.

---

Conclusion

The Raise Summit, as captured in this Sourcery episode, is a microcosm of the AI ecosystem: fast-evolving, opinionated, and dizzyingly multi-faceted. The fundamental takeaways: interfaces are shifting from screens to speech, open source’s role is surging, chips and deployment are diversifying, and the stakes—in both economic and societal terms—are enormous.

For anyone tracking the future of AI (or just wondering where their mouse and keyboard will be in five years), this episode is essential listening.

Transcript

[Molly]

Raised, raised, raised, raised.

[Max Cook]

Paris, Paris, Paris, Paris.

[Ramin]

Paris is beautiful, but it's hot as

[Molly]

Paris in the middle of the summer, middle of the heat.

[Max Cook]

People are ignoring it.

[Molly]

They're showing up. Okay, we have a very special guest. I know I've said that before, but we really do this time. Coming off the heels of the Jamon interview at Kotu, we have Max Cook of CO2. Max, welcome.

[Max Cook]

Thank you. Excited to be here. It's pretty nice.

[Molly]

It is very nice. In this very secret location in Europe, which we cannot talk about, not disclose

[Max Cook]

the official location, but you've got an old cement wall which you have many of in Europe.

[Molly]

So it might be anywhere. It could be anywhere. It could be a country that Philippe and Thomas know very well.

[Max Cook]

True. That is true.

[Molly]

Cannot confirm, though.

[Max Cook]

No.

[Molly]

Okay, so we have to get into this. I think you have a couple of big insights you want to share with us today.

[Max Cook]

I hope they're insights, but yeah, I mean, I have a couple things I've been thinking about.

[Molly]

Like, what's your hottest take?

[Max Cook]

Hottest take might be that the keyboard and mouse are slowly dying. I think.

[Interjector]

Damn.

[Max Cook]

I think Code two. You've heard Jamin speak about it. You've heard Philippe and Thomas speak about it. We talk about big tech waves. The you went from mainframe to PCs to networking PCs to desktop Internet, which is where you get a mouse and keyboard, to mobile Internet, which is where the iPhone supremacy began. And with each of these big waves, we end up interacting with compute, interacting with computers in a different way. And it seems pretty clear at this point, I think there's a good stat that, like, the Average person has 60 to 80 apps on their phone.

[Chris]

Wow.

[Max Cook]

They use 11 apps daily on average, and 30 apps in any given month. Like, it seems quite clear that we're going from a world in which you have 60 to 80 apps to an agentic AI world where you've got a lot fewer than that. 2, 3, 4 apps, where everything else plugs in as an API or an mcp. And in that world, it would only make sense that we reinvent how we interact with compute. It's not the keyboard and the mouse. It's processing probably natural language. Like, if we're talking, we speak in natural language. If you're working with a coworker, you ask them what to do, you figure it out together, and you go do it. So I think there's been a number of good discussions about this. Michele from Replit was talking about how natural language is how people communicate. That's clearly the direction we're going in. There's a Greg Brockman quote that I always think about. I'll butcher it. But he had a tweet about, we've always conformed ourselves to computers, like with data, by inputting specific data with a keyboard and a mouse. And now we're finally entering the era where computers have to conform to us. And so this microphone is reminding me we're already seeing it at CO2, like in the famous CO2 bullpen that you got to see. We've got a number of investors that are whispering into their collar half the day, really. Because that's how to give models better context, better prompting, and ultimately, like, produces much better work output. So that's my. That's my take. The keyboard and mouse, you should throw them away. And we're going to. Natural language interacting with agentic models.

[Molly]

Sounds like bullish on humans.

[Max Cook]

Bullish humans.

[Mark (CTO of AMD)]

Yeah.

[Max Cook]

No, I agree. Yeah. I think, sure, we're building for like an autonomous agentic future, but also there's going to be a lot of that which is just supercharging humans. Like, that's what we see at CO2 is all of our analysts, all the investors that are whispering into their collar are producing better and more work output because they're able to more fluidly interact with these models. I think the Thinking Machines interaction model demo that came out in May kind of went under. Loved. There wasn't enough hype around that.

[Molly]

Really? Why say more?

[Max Cook]

Yeah, I mean, to me, that Thinking Machines interaction model demo was like a bit of a peek into the future. You're going to interact with compute at all times. It's got to have zero low latency back and forth. It has to be able to input images, understand what you're saying in real time, respond to you. And so I watched that and said, of course, this is like going to sit on top of every intelligence layer, every model, every app. That's how we're going to interact with compute. And I think people were excited about it at the time, but haven't necessarily put the dots together between speech to text interaction models and thinking about the historical context, which is that every single time we've moved from one era of tech to another, the device has changed and how we interact with that device and with compute has changed. So I think it's only logical that that's happening again now.

[Molly]

Damn. Also sounds like you need assembly AI speech models.

[Max Cook]

Definitely. Yeah, I know there's. Assembly's working on it. There's a lot of exciting companies that are working in this space and I think we'll see what the device ends up looking like, but I think there will be a new format, a new way to interact with AI versus how we used to input data into a structured output via graphical user interfaces, GUIs, apps, keyboards, mice. Like to me that we're going to look back in five years and say, of course we're not using keyboards and mice anymore. We've come up with a better way to interact with intelligence.

[Molly]

All right, so we'll upgrade your wires to neuralink chips.

[Max Cook]

Yeah, maybe that's it. I mean, obviously wearables is a place people are spending a lot of time. Neuralink is a place people are spending time. I don't know what the answer is, but like, to me it's a hell of a lot easier to see that, like the keyboard and mouse are dying a slow death. I can't tell you. It's for smarter people than me, builders, operators, founders, to figure out what the new format looks like. But I think that has to be the direction we're going. Where you've got OpenAI coming out with a super app for the first time that's going to introduce almost a billion weekly active users to agents for the first time. It's going to change how people interact with their phone. It change how people want to interact with their phone. And we're going to need something new for that.

[Molly]

And probably some MongoDB databases for all those apps.

[Max Cook]

Yeah, definitely. I mean, you've seen it. You've seen it in some of the public software companies already. Private too. Snowflake, Databricks, Datadog. All of these businesses that are tied to growth in AI app production, AI app development are doing great. And Mongo is going to be the exact same type of story.

[Molly]

Amazing. Well, Max, I got one more. You have one? Just wait, there's more. What? What is the second hot Take one more.

[Max Cook]

Just because I've been spending a lot of time on it. Yeah, this one I don't have an answer for. So I'll just leave you with a question that I don't think people have thought enough about.

[Molly]

Okay.

[Max Cook]

There's a lot of discussion of open source models versus closed the frontier, OpenAI and anthropic. Right now, a lot of the discussion I see is sort of there's either one is a winner and one is a loser, the other is a winner and the other is a loser. I think that's totally the wrong way to think about this. I also think there's been some great posts. Just Jesse Zhang from Decagon Posted recently on X about how I don't want to get his numbers wrong, But I think 90% of Decagon's workloads run on open source models.

[Molly]

Damn.

[Max Cook]

And that is because they are at kind of production scale. They need low latency. They need to be able to fine tune a model and use a small model for that low latency. That's not a product that Anthropic or OpenAI are offering today. You, you can't do that with whatever their Frontier model is. So they're on 90% open source. Yet at the same time Enterprise LLM spend was 19% open source last year and it's 11% this year. So it's dropped. Yet you've got businesses like Decagon telling you that 90% of their workloads are done via open source. Jesse's explanation I think is the correct one, which is that open source is good for production grade production scale workflows. Yet we're so early in this AI journey that there's not that many production grade workflows. And so if you are trying to discover a new use case, you want the smartest model you can possibly get, you use the Frontier. And I think the question that I don't think is being asked in this open source, closed frontier debate is clearly people want to do model routing. People clearly want to fine tune and customize smaller models. I don't think anyone's asking the question of what happens when Anthropic and OpenAI decide we can offer that as a product. I don't know that they're going to do that, but it seems plausible that they would see this opportunity to take their older, cheaper, more efficient smaller models and sell that as a product.

[Molly]

I wonder if OpenAI will open source.

[Max Cook]

Yeah, I don't know. I mean as of now they won't. But model routing I think will be in their best interest, maybe fine tuning of some kind. So that's like the question I don't think is being asked enough. When we consider what does the end state compute ecosystem look like? X percent open, X percent closed. I don't think people are thinking about the idea that like a Microsoft who's saying they can help you build your own model and fine tune that, then Anthropic and OpenAI could just as feasibly offer that product at some point.

[Molly]

Damn. Well, we just had Matan from factory on and he said within his enterprise customers he saw in the beginning of the year it was less than 1%, got to 1% in January, then I Think it was this month or something like that. I could be quoting this completely wrong, but it got to 10% open source.

[Max Cook]

Yeah, I mean you're seeing it on X, which is obviously one of the better places to see some of the people in the forefront talking about what they're doing. I don't think that's a crazy end state of the world. End state of AI compute. AI inference would be, I don't know, 10% open source. The Decagon Jesse's article made a ton of sense. If you know all of the parameters of the workflow you're trying to execute, you've thought about all the edge cases, you've used a frontier model to hone that, then it makes sense that maybe you're actually seeing better capability with an open source model that's fine tuned or the combination of a frontier model for your hardest tasks and an open source or previous version of a model for your kind of production grade workflows. Like customer service is a great example of that. But to me that's. We're really just so early on in terms of what will be the kind of adopted production grade workflows that are built on top of these AI models that the frontier is. I mean everything's growing. And so I think that this kind of zero sum discussion of open versus sourced is missing the point. And there are a lot of questions that I don't think people are fully thinking about yet. But it's interesting. Like it's fun to be sitting in a world that we don't know what the end state looks like. But it's our job to at least try to predict what it could be.

[Molly]

Amazing. Two hot takes. And I think where we are, which is not a very large palace in a very random European country, is a good kind of. It's a good symbolism of how far AI has gone. We can now buy palaces.

[Max Cook]

True, true. Yeah. I mean this is just the early

[Pim]

days of the AI.

[Molly]

This is just the early days. This is literally just the early days. So, okay, speaking to that and raise, what are you most excited about with raise this year?

[Max Cook]

Yeah, I mean I think the lineup of speakers is amazing. That cuts across like the compute layer, infrastructure layer, app layer. There's investors, there's venture investors, public equities investors. And so to me, to have all these people in one place to be able to talk about these questions that I'm asking around open source frontier models, sovereign AI power constraints, grid bottlenecks, that's what I focus on all day, every day with my colleagues at CO2, but it's really, really damn fun to get to talk about it with founders, operators, entrepreneurs, and hear what they're thinking on these same topics. So having all these people in the same place is what I'm most excited for.

[Molly]

Amazing. Great place to end it. Thank you so much, Max.

[Max Cook]

Yeah, thanks, Molly.

[Molly]

The sun came out for Henri. Henri, thank you so much for having us here in this very secret location. And I think it's. We're in Germany, is that correct?

[Henri]

Exactly. We are in south of Germany. Thanks for being here, Molly. It's going to be a fun night. Thanks for being here and welcome.

[Molly]

So tell us about Reyes this year. It's huge and very fancy.

[Henri]

Yeah, it's fun. The idea was to bring the German savoir Fairfair. So we're not in Germany, we're in France. We can't really disclose where we are. That's part of the night and the whole deal. But it's kind of bringing a very different iconic experience to a very tech event. So we are all in our tech bubble, token maxing on a daily basis, but it's good to have a bit of a history in our north side. Right? Thanks for being here, everybody.

[Molly]

Of course. I know you're the most popular girl at the party right now, so we're going to keep this really, really short. What is your hottest take right now?

[Henri]

So my hottest take is Germany is actually extremely beautiful and we're super, extremely fortunate to be here.

[Molly]

Perfect. Thank you, Henri.

[Henri]

Thank you, Mari.

[Philip Johnston]

Have a good night.

[Molly]

We're here with Pim of General Intuition. Pim, welcome to Sorcery.

[Pim]

Thank you for having me.

[Molly]

Okay, so for those who don't know what general intuition is, can you give us a little brief overview?

[Pim]

As humans, we can talk or choose to sort of output text as one of the actions we can take, whereas an LLM is kind of the only thing they can do. So we train models that can predict lots of different actions where text is just a subset, which is a much more general approach in LLMs.

[Molly]

And what is your hottest take today?

[Pim]

Most of AI is just glorified data science.

[Interjector]

Damn.

[Molly]

Say more.

[Pim]

Good models mostly are downstream from good data and simple things.

[Molly]

So what are you most bullish on in that respect?

[Pim]

Really good data sets for the most part.

[Molly]

Where do you store that data?

[Pim]

I cannot disclose that.

[Molly]

Well, if you need to store your data, put it in MongoDB databases.

[Max Cook]

No, no, no, no, no.

[Pim]

Okay.

[Max Cook]

All right.

[Molly]

Any other hot takes?

[Pim]

Yeah. Yann Lecun is underrated.

[Molly]

You think Yann Lecun is underrated. Why?

[Pim]

The friends just told me to say that I pulled into a room and they were like, you have to say this.

[Molly]

No more comment. He's gotta walk away now. Damn. Okay, see of him. Oh my God. Hi. Okay, we're here with Mark, CTO of amd. Mark, welcome to Sorcery.

[Mark (CTO of AMD)]

Thank you, Molly. Glad to be here.

[Molly]

Amazing. We're in a very secret location in a European location. I just. We can't. You don't know where we are, but it's very secret. Okay, so I have to ask you, we're at the Raise summit. AI is booming. What is your hottest take right now?

[Mark (CTO of AMD)]

Hottest take is people don't even understand how transformational agentic AI is. It's just transforming how work gets done. You know, just entire end to end workflows are being reimagined. And I think it's starting to come out when you hear some of the stories this week at Raise.

[Molly]

And what is your biggest lesson that you learned in the last maybe two or three years of this boom at amd?

[Mark (CTO of AMD)]

Well, amd we've, we got ahead of it. We've worked getting our CPUs and GPUs ready for this and now with these Agenta workflows, you actually need both. And fact, the ratio of CPU to GPU is becoming like one to one. But none of us anticipated how fast these agentic workflows would just take off. The last six months have just been amazing and it's just a harbinger of what's yet to come. We're just at the early stages.

[Molly]

How many agents are you using today?

[Mark (CTO of AMD)]

I. Every day I start with agents, they run my day, look at what's coming up. But what's really cool is how we do our chip. Our chip design. We have thousands of sub agents just doing incredibly complex chip design, speeding how fast we can get the next generation to market.

[Molly]

Amazing. Okay, as the drums roll, what are you most excited about with Raise this year?

[Mark (CTO of AMD)]

Raise. Well, it starts with the President of France and ends with Yann Lecun. I think we're gonna. We got it bounded pretty well here. No, it's great. It's really great speakers. I think we're gonna really hear some insights across every facet of the AI ecosystem.

[Molly]

Amazing. Well, thank you so much, Mark. Really appreciate it.

[Mark (CTO of AMD)]

Thank you, Molly.

[Molly]

Okay, we're here with Robin of Black Forest Labs again at a secret location in Europe. Robin, welcome to Sorcery.

[Robin]

Thank you. Hi, nice to meet you.

[Molly]

Nice to see you. I saw you a Couple months ago.

[Robin]

Really?

[Rodrigo]

Where?

[Molly]

In London, at the Legends Gallery.

[Robin]

That's right, yeah.

[Molly]

Tough night. Anyways, okay, so for people who don't know what is Black Forest Labs, we

[Robin]

are building multimodal visual models for content creation and for now, physical AI.

[Molly]

Amazing. And so we were talking about a couple hot takes earlier, but I'm really curious, what is your take on open versus close? Open source versus closed?

[Robin]

I mean you can approach this from like so many different angles, but I think like the most important one is that open innovation is good for the world. And I think fear mongering around AI models is only going lead to them being more close and that's ultimately going to slow, leading to slow down all AI progress. And I think ultimately it's super important that AI is actually an accessible technology. It's also going to make it more safe in the end. I think that's like something that is a pretty fundamental principle. If you don't think about like business models that you build around open versus closed source and all of that, then you can start to be more like elaborate. But I think fundamentally open innovation is good for the world. Is it open weights? Is it open source? It doesn't really matter in the end, but I think having a accessible AI models is key.

[Molly]

So you're based in Black Forest, Germany. We're here in Europe and there's a nice convergence of European tech and American and Silicon Valley. Everything that's converging over here. What should people know that aren't from Europe about what's actually going on in European tech right now?

[Robin]

I think it's getting better. There's not enough going on yet, I would say. Even though I don't know it's like a fancy event like this, it looks really nice. Right? But I think what really matters in the end is that Europe is not known for having, I don't know, nice castles and stuff like that, but actually nice AI or nice technology. And I think that's something that, yes, it's changing and I think events like that where a lot of people come here, I think that's a good start. But ultimately there needs to be like a fundamental mindset to really push AI innovation from here as well.

[Molly]

I know you're super excited about the raise summit, so what makes you so excited?

[Robin]

I'm not going to answer this question.

[Molly]

Seeing all your friends maybe?

[Max Cook]

No, it's great.

[Robin]

Yeah. I mean there's like so many good people here. I think it's fantastic.

[Molly]

Perfect place to end it. Thank you so much, Robin. Great to see you.

[Philip Johnston]

Thank you.

[Interjector]

This episode is brought to you by Brex. My favorite. You become what you spend on and I refuse to spend my time on work that shouldn't exist. Expense reports, receipt chasing and manual closes the companies building. What's next from Vercel, OpenAI, Anthropic Granola and Deepgram all made the same call. They all run on Brex. BREX is the intelligent finance platform that combines cards, expenses and banking into a single stack with agentic finance built built in AI agents that handle expenses, automatically enforce policy before spend happens and close your books in minutes. That's why sorcery runs on Brex so I can spend time on building and not busy work. It's time to get Brex AF. Learn more at brex.com sorcery that's B-R-E-X.com S O-U R C E R Y Bye.

[Molly]

We have a very special guest here today. We have Matan Factory. Matan, how are the droids doing?

[Matan]

The droids are doing well. The factory is growing very excited on things that have been happening.

[Molly]

Are the droids in Europe with us right now?

[Matan]

The droids are in Europe. They recently made the jump across the pond. A lot of the largest banks and enterprises in Europe are now automating their software engineering with Droids and with factory. We're very excited about it.

[Molly]

So we have a very difficult question for you today. What is your hottest take?

[Matan]

My hottest take is that in the next 12 months, 90% of tokens will be going to open models. Now, not all the tokens and maybe not the most important tokens, but I think, you know, Big Token has been spewing some propaganda against the Chinese models, which is, you know, how they're labeling the open models. And I think in fact these open models are incredibly performant, incredibly cheap, incredibly fast and there is going to be an important use for them in kind of the portfolio of models. I think still the frontier closed models will have a place, but I think that place will be shrinking, at least in token share. Maybe not in cost, but in token share.

[Molly]

Are you measuring the switches on your side?

[Matan]

Yes. And at the beginning of the year we saw enterprises were using less than 1% open models, like less than 1% of their tokens were going to open models. Then in around March it kind of crossed that 1% threshold and then in May it crossed the 10% threshold. So it's growing pretty quickly because of the kind of cost optimization and also like, you know, not every task you need Opus 4.8 or like GPT 5.6 ultra high. And many tasks day to day you do are very simple. And you know, at factory we will actually dynamically route you to whichever kind of open or cheaper model is better for that task.

[Molly]

What's your prediction for percentage of open usage by the end of the year?

[Matan]

By the end of the year, I think we'll probably get to. At least in the enterprise, we'll probably cross the 50% threshold by the end of this calendar year.

[Molly]

Oh, my God. Breaking news. This is crazy. Okay, as we finish up, anything else you want to share? What are you most excited about for raise?

[Matan]

I'm most excited, honestly, about AI making its way into Europe. They've been, you know, they haven't been as urgent about it, but I think this is kind of a sign that like AI is making its way across the pond. People really care about it. There are a lot of businesses that can benefit dramatically from some of the kind of aggressive, sometimes too aggressive moves that people have been making in the United States. Excited to see more of that urgency in Europe and there are a lot of great companies that it's fun to work with here.

[Molly]

Amazing. Thank you so much, Maton.

[Matan]

Thanks for having me.

[Molly]

We have Philip Johnston of Star Cloud, one of the hottest companies in space right now. Philip, how's it going?

[Philip Johnston]

It's going great. Thanks very much for having me on the podcast.

[Molly]

How does it feel to be in SpaceX's S1?

[Philip Johnston]

Amazing. I mean, to be honest, I was not expecting that. And the article they quoted me in, to be honest, I. I'm very surprised they picked that one to quote me. But yeah, it's.

[Molly]

What did they quote?

[Philip Johnston]

So I had to write as part of getting, as part of doing the World Economic Forum, like Davos thing, you have to like write an article. And I basically very rapidly threw together an article on data centers in space, I guess because it was wef they like picked that one. But yeah, I'm surprised they picked that one.

[Molly]

It must have been really good. Okay, so we have to ask you, we're here on Earth, but what is your hottest take right now?

[Philip Johnston]

My hottest take right now is that the SpaceX IPO is going to be viewed historically as the most undervalued IPO of all time. I think they will tear through 10 trillion within, you know, certainly within the next couple of years and then have almost unlimited TAM for what they're building. So in historical terms that we viewed incredibly undervalued.

[Molly]

This is not investment advice, but why are you so bullish?

[Philip Johnston]

I'm not like regulated by anybody. So I kind of tell people, just buy this basic stock market. It's a great stock to buy. No, but it's mainly because so space not just because of orbital data centers which we're doing, which lots of other people are doing. They own what will be by far the most cost effective launch vehicle. And then that opens up every industry in space that will be possible beyond that. So. So that's all of asteroid mining, lunar resource mining, all of the comms businesses that are going to be built. Everything else is going to have to go through SpaceX. Yeah, it's like owning the railroads. Lots of businesses will be built on top of it. Like our business will be built on top of it. But the railroads are a great business to own for sure. Yeah.

[Molly]

So there are a handful of chip companies here this week. We're speaking with Cerebras Sombinova. We just spoke with the CTO of amd. Do you have a favorite chip?

[Philip Johnston]

We welcome all chips. We're actually going to be flying. We're speaking with Samanova. We hope to fly the salmonova chip with Cerebra 7, an incredible inference chip. We have to fly their chip within Nvidia. The like Grokkixture actually is, you know, can be potentially well suited for space as well. We flew an arm ship on our first spacecraft. We threw three arm GPUs on our first spacecraft. Nobody knows about that, but we did. So yeah, we are chip agnostic.

[Molly]

And so how's the, the factory going? How's is it? Are you guys outgrowing it already? What's going on?

[Philip Johnston]

Yeah, so we are about to move into a much, much larger facility. We're building a kind of campus to set up a huge manufacturing line for what we're calling StarCloud 3, which is like the vehicle which will fit on Starship. So we can potentially fit up to 10 megawatts of compute capacity per Starship launch. And so yeah, that's in August. We're going to move in and you will be the first to do a tour of it.

[Molly]

You heard it here first. We're going to do a tour. Okay, so as we close out, what are you most excited for for this week?

[Philip Johnston]

I mean they just have some incredible speakers and incredible other attendees. So yeah, I'm excited. I've never met the CEO of Samanova. The AMD CTO would be great to meet. So yeah, excited to meet some people.

[Molly]

We just recorded an interview with Rodrigo of Samba Nova this morning. Fantastic. Really recommend that interview. If you haven't listened, if it maybe it's out yet, I don't know if it will be. But he is fantastic. Fantastic at explaining everything. Yeah. Okay, Philip, thank you so much.

[Philip Johnston]

Thank you so much.

[Chris]

Oh, my God. You guys have to see what's happening at this undisclosed location.

[Molly]

All right, we're here with Chris of Good Future Media.

[Chris]

Chris, welcome to Sorcery Massi Beaucoup Molly, very nice to see you here at the.

[Molly]

We're gonna bleep that one out because we don't know where we are right now and we're in a secret location, but Sava.

[Chris]

Well, this is a nice return because I used to live in France for four years, so AI France together. I had to come to this event.

[Molly]

Really? What were you doing?

[Chris]

I was an English teacher before, actually. And so for 10 years I was teaching in the classroom. And then around the pandemic, I started listening to podcasts, the all in podcast, to be specific. And I was learning so much and I knew that I had to teach the Internet about what I was learning on the all in podcast. And I started clipping up the all in pod. And that's kind of how I shifted into this new business of clipping. And so now I have a huge agency. We get out thousands of clips a month. And yeah, it's amazing what clipping has done to the social media environment. And it really is the game nowadays.

[Molly]

Clipping is the new ad.

[Chris]

Clipping is a new ad. It's a way to get yourself out there. Whatever the long form is, whatever message you have, whatever idea or thing you're trying to sell, you need to get into a short amount of time to be able to grab somebody's attention and get them to maybe check out your longer form content or what it is you're trying to sell or pitch.

[Molly]

Okay, I have to ask you for my own benefit, what is the secret to clipping?

[Chris]

It's really the first three seconds. The three seconds need to be grabbed the viewer in. If the spoken line isn't catching you, you need to have a visual hook on the screen that will tell the viewer what is the payoff for them. What's the value that the viewer is going to get from watching this clip? If you don't have that in the first line of the speaker or a hook on the screen, think, you know, restart, think of something else. Because it's not going to go viral. It's not going to do well.

[Molly]

What is the most well performing clip you've ever made?

[Chris]

We have Chamath Talking about his, you know, newest startup going. I think we got 4 million views for him. David Friedberg as well, warning about the famine in Ukraine at the start of the Ukraine war. That one went wild on TikTok. Five million views. And, you know, really the all in besties have been an inspiration for us for getting lots of good content out there. Worked with Harry Stebbings at 20 VC for two years, and now we're working with venture capitalists and CEOs and investors across the Silicon Valley landscape.

[Molly]

What is the price? What is the standard price of clipping today? Yeah, I think I get fleeced. Well, quite often.

[Chris]

It. It is more expensive than you'd think when you're used to AI clips, like, there's a lot of AI companies out there that will, like, take your podcast, look at the transcript, like, pop out five of the best clips and put captions on there and call it a day. But you really need the human thought, the human psychology going into the clip and adding other visuals to keep people on their toes while watching. And that takes human time and skilled human time. So, you know, 200 bucks a clip is pretty standard in our agency. You know, minimum. Then you want to go clips daily. So a clip a day at least. If you can do two or three a day, that's even better. So that kind of starts to get you some of the pricing for a better company like ours.

[Molly]

Okay, if we were gonna make this clip go viral, what should the hook be?

[Chris]

Oh, my God. You guys have to see what's happening at this undisclosed location. You won't believe who's here. There's undisclosed names and Molly o', Shea, Shannon.

[Max Cook]

And I messed up.

[Molly]

O' Shea is.

[Chris]

And I messed up the host name.

[Molly]

Oh, my God.

[Chris]

And I can't wait until you see what's next. Something like that.

[Molly]

All right, we'll lead with that. Chris, thank you so much.

[Chris]

Thank you. Thank you. It was very nice to meet you in person. I'm a big fan of your show.

[Henri]

Thank you.

[Molly]

I appreciate it. And I appreciate any clips that you guys just happen to make.

[Chris]

Hey, I think we've made some of your clips, actually. The Techno Optimus prime is our Instagram account, and it's a niche channel where you just clip up our favorite people talking about tech and AI. And you're on there too, so.

[Molly]

Amazing.

[Pim]

Yeah.

[Molly]

Yeah. Awesome. Well, thank you so much, Chris.

[Chris]

Thank you.

[Interjector]

If you're building what's next in AI, you need to know MongoDB, the database platform developers love and built for the agents you're running MongoDB stores, searches and reasons over your data in real time with vector search and embeddings from Voyage AI all in the same system. No separate pipelines, no stitching together 10 different tools. It's why 75% of the Fortune 100 and leading AI native startups run on MongoDB. Build and scale from your first user to billions of vectors. Go to mongodb.com AI to learn more. This is. That's mongodb.com AI to learn more. Bye.

[Molly]

We are here with Ramin of Liquid AI. Ramin, welcome to Sorcery.

[Ramin]

Thanks for having me.

[Molly]

So we're in a very secret location that has no ac, but we can't say anything more than that. What are you so excited about? Or why are you so excited about rays this year? And tell us about Liquid AI.

[Ramin]

Definitely. So I feel like this year they have exponentially kind of improved the quality of their guests. That's what I can say. And then the structure is much more mature as I've seen it before. There's multiple conferences happening at the same time. There's Makina and then there's Raise. And everything is related to the stuff that we do. We are bringing foundation models to the physical world. So physical AI is extremely important for us and that space is becoming more and more interesting. We're building foundation models that are so cheap that you can bring them on Raspberry PIs, for example, so you can host them on any kind of device that is on the planet.

[Molly]

So you're saying it's really hard to token max with liquid AI, correct?

[Ramin]

Yes. I mean, tokens are basically like, we're bringing the cost of tokens like to zero, you know, and we are just bringing the value of the AI for enterprises, basically. So like imagine if, if there's not tokens that are actually important, it's just the, let's say the outcome that actually matters. And this has been like now, I think the discussions in enterprise AI, which is like one of the hot takes, I guess.

[Molly]

You know, Carp has something to say about that. I don't know if you saw him.

[Ramin]

Yeah, I saw him. That was incredible, you know.

[Molly]

Any notes?

[Ramin]

Yeah, I mean that I, I, I fully agree with him, you know, like 85% of everything that is happening right now in AI goes into research and development of code.

[Chris]

Right.

[Ramin]

And then we don't know what this code is going to do, you know, And I like that he said the jig is up, you know, because enterprises like now, in fact actually in Europe, they are actually waking up right now, you know, and they're getting into the place where they really see they have to take their time, you know, similar to how they have been treating like any other technology to really understand the value of a tech and, and then deploy basically those solutions into their products. For example, one of the things that we did, we are powering in car intelligence for Mercedes Benz. So foundation models, getting inside the car and making your car basically smarter, this is what we do. And then this is done by a traditional oem. You can imagine Mercedes Benz, one of the flagship brands in Europe, but, but at the same time they opened up to a technology that can actually truly unlock value, you know, and I feel like we have to, in enterprise AI, we have to start thinking as foundational companies. We have to start thinking about that value chain a lot more. Which I think he was on the right, right path.

[Molly]

Yeah. So we have to ask you, what is your hottest take right now?

[Ramin]

Well, all foundation model companies from 20, 20, 19, when, let's say pre training started like panning out, you know, like we saw that we have to kind of, you know, scale these AI systems, they started working on one axis. You know, the axis was maximizing intelligence at all cost. You know, now we are at a place where we see that look, efficiency is not an afterthought, energy is not abundant. You know, chips are, you know, like there's not that many chips like available in the world right now. You know, like you have to invest and also like supply chains are getting disrupted on the memory side of things. Government can have control over your assets, you know, like, and whatever you're building and safe rollout of artificial intelligence at the largest scale. If you have maximize intelligence at all cost, that access alone is not going to get you to the place that you want to go. At the same time, open source is also becoming like more and more kind of popular popular, you know, as you're seeing in enterprise AI, especially like as these governance issues are coming in. So I always wanted to propose like this three axis problem, you know, instead of one axis problem, imagine you're maximizing intelligence but at the same time while you're designing a foundation model, care about its cost. So at the same time like you're not just going to push transformers out, you're going to like really work on, let's say like finding out what is the best, what is the best computational graph of intelligence that runs on a certain kind of processor that you want to run it. And then so efficiency and cost of intelligence as A, you know, like a first class, first class citizen, you know, and not afterthought, you know, because usually now AI companies right now are thinking about doubling down on efficiency of their models, distilling their models down, you know. But this is not something that is, you know, like we took it. So we actually took that axis of efficiency very, very seriously. That's axis number two. The first one was capability as intelligence. The third axis, I would say substrate. Where does this intelligence system go? You know, so if you, if you think about AI, AI is majorly like getting hosted in data centers. But you could also bring intelligence on phones, on laptops, on airplanes, on cars. You know, like you can actually host AI at any different substrate. Right? And this is something that I think if you co evolve it with the capabilities of the models, you can have like true enterprise AI capabilities.

[Molly]

Damn, it's a hot take.

[Ramin]

Well, it's also very hot here.

[Molly]

It is very hot here. Europe or wherever we might be really needs to invest in AC and probably cloud seeding. I think Rainmaker might be able to help with these droughts and the heat, heat wave. Okay, any, any last thoughts? By the way, this segment of raise is sponsored by MongoDB, Rex and Assembly AI. So if you need a database, money, help or text to speech and speech models, that's what you get anyways. Okay Ramin, to close out, any last thoughts, what are you most excited about?

[Ramin]

I'm actually excited. Tomorrow I will be on a panel with Ian Stoica who is like a legend of the field, a co founder of Databricks and co founder of like Arena. Like we are going to actually kick off the conference tomorrow, you know at 9am and excited to tell the world about open source and you know, the impact of like let's say like these three axis that I actually talked about into absolutely realizing kind of the value for enterprises, you know, like not just thinking about like foundation models as like token machines that are generating money and revenue for the famous amount of companies that are useless tokens. Right. And getting them into the place where they can actually unlock value for enterprises.

[Molly]

Amazing. Trees are good. Thanks Ramin. Thank you Rodrigo. Welcome to Sorcery.

[Rodrigo]

Thanks for having me.

[Molly]

We are in Paris, we have a long interview coming out soon. But before that you have some really big news today. So what happened now?

[Rodrigo]

We're super excited. We're announcing the first close. Our series F is a $1 billion raise at 11 billion valuation.

[Interjector]

And who led this round? Who are some of the investors in this one?

[Rodrigo]

The round was led by General Atlantic. Really excited to have this incredible investor on board and significant investment from Seligman Ventures, from T Rowell Price and Capital Group.

[Molly]

What are you going to do with this money?

[Rodrigo]

You know, we're seeing incredible interest in our, in our product and so we're using this capital to accelerate our supply chain and making sure we can deliver the racks to all these, all these customers worldwide.

[Molly]

So you have a chip with us in the room today. Can you please show us the chip?

[Rodrigo]

This is it. This is the exciting new product that we're shipping later this year, SM50. You know, it is a cloud scale RDU that allows us to actually deliver premium inference at a significantly higher performance that you can get with GPUs at a fraction of the cost, a fraction

[Molly]

of the power for people who don't

[Interjector]

know what premium inference is. Can you explain that a bit?

[Rodrigo]

Yeah. So the world of inference is running into an economics problem. And so what we're able to do with premium inference is drive the performance of the largest models up so you can actually deliver high quality models, these trillion parameter models and at their full precision at incredible speeds. And when you're running faster with the high accuracy models, you can charge more at a lower cost, allowing their service providers to generate more margins for their services.

[Molly]

All right, so last question before we go and then you have to check out the full interview afterwards. But what is your hottest take?

[Rodrigo]

Well, you know, the world's moving into inference and so we're excited to be here. We're excited to actually have a new trip to share with the world. And, and I think you're going to see Samonova be part of a number of these services, you know, before the end of the year.

[Molly]

Amazing. Bullish on Samba Nova, by the way, how did you guys come up with that name?

[Rodrigo]

Well, you know, I, I grew up in Brazil.

[Molly]

Okay.

[Rodrigo]

Yeah, I grew up in Brazil and, and Samova is a new dance. It's a new dance, it's a new trip. And so check it out.

[Molly]

Wow.