The Race to Run AI From Space

Summary

WSJ Tech News Briefing

Episode: The Race to Run AI From Space
Date: December 12, 2025
Host: Katie Dayton
Featured Guests: Ellen Gammerman (WSJ reporter), Micah Madenberg (WSJ space reporter)

---

Overview

This episode of the WSJ Tech News Briefing delves into two bleeding-edge developments in technology:

1. The controversial rise of AI-generated actors in Hollywood, centering on the debut of Tilly Norwood—the industry's first fully AI-created movie star.
2. The ambitious push by tech moguls to move data-heavy AI infrastructure beyond Earth’s atmosphere, with companies like SpaceX and Blue Origin planning to run data centers in orbit.

Both segments highlight the rapidly advancing intersection of artificial intelligence with creative industries and global infrastructure, raising urgent questions about human replacement, technical feasibility, and the future of tech innovation.

---

Segment 1: Tilly Norwood—Hollywood’s First AI Actress

[00:19-05:46]

Key Discussion Points

• Introduction to Tilly Norwood

  • Tilly is a brown-haired, brown-eyed, 24-year-old AI-generated actress, intended as Hollywood’s first AI movie star.
  • She was created over six months and has already sparked controversy. Her persona is described as sassy and oddly self-aware (01:41).

• Her Creator: Aline van der Velden

  • London-based, former actress, and head of an AI production studio.
  • Created Tilly after seeing the emergence of AI influencers and predicting AI movie stars as the “next thing” in entertainment (02:29).
  • Van der Velden’s background includes making comic videos for YouTube and BBC3.

• Personality and Public Perception

  • Tilly is still in development—her “brain” and personality are actively being built to enable future one-on-one interactions with fans (01:41).
  • A notable quote from Ellen Gammerman:

    “She’s a little bit sassy, but she’s also a little bit weirdly self-aware. Like she calls herself the end of civilization because so many people have been so upset about her being created.” [01:47]

• Industry Backlash and Fears

  • Tilly’s announcement met immediate criticism, notably from director James Cameron who reportedly called her “horrifying” (03:18).
  • Concerns center on the potential for AI to render traditional jobs obsolete, now extending from technical to creative fields (03:35).
  • Quote from Gammerman:

    “It’s tapping into these fears that everyone there is replaceable, including now actors… this could start seeping into the creative arena.” [03:35]

• Aline’s Response to Criticism

  • She sees AI film as a separate creative domain, not aimed at replacement but expansion:

    “AI can build worlds in ways that traditional filmmaking can’t… there’s a kind of a creative renaissance.” [04:00]

  • Many in the industry remain skeptical about coexistence without competition.

• Hollywood’s Secret Interest

  • Van der Velden has signed more than 60 NDAs with Hollywood “players” interested in AI or hybrid films; current projects are cloaked in secrecy, reflecting industry caution (05:03).

Memorable Moments

• The suggestion that Tilly “calls herself the end of civilization” (01:47) encapsulates the satire and unease surrounding Hollywood’s AI evolution.

---

Segment 2: The New Orbital Data Center Space Race

[06:35-12:05]

Key Discussion Points

• The Premise: Why Move Data Centers to Space?

  • Rationale includes ground-based challenges like supply chain delays, high construction costs, and massive energy needs (07:16).

  • Proponents point to abundant solar energy in space and falling launch/component costs making the idea more feasible than ever before.

  • Quote from Micah Madenberg:

    “The pull factor is literally straight out of science fiction… we’re at the point where launch costs and satellite component production costs maybe are low enough to make that science fiction dream a reality.” [07:48]

• Technical Vision: Thousands of Smaller Satellites

  • Rather than few giant data centers, the concept involves launching “thousands upon thousands of satellites” in networked constellations (08:24).
  • Smaller satellites linked as clusters are argued to be more practical for data transmission than monolithic structures.

• Cost and Feasibility Challenges

  • SpaceX’s Starlink currently operates the world’s largest satellite network (~9,000 satellites), illustrating the difficulty in just achieving the scale required (09:01).
  • Even with declining costs, launching and maintaining vast networks would incur “costs that pile up really, really quickly” (09:01).
  • Further, uncertainties about terrestrial improvements (new energy sources, easier construction) mean ground-based AI data centers could remain competitive.

• Who’s Investing, and How Seriously?

  • Google and Planet Labs are planning an AI compute satellite experiment for 2027 (10:19).
  • Blue Origin (Bezos) and SpaceX (Musk) are leading major, ongoing efforts.
  • Even Sam Altman, OpenAI’s CEO, is reportedly seeking rocket partnerships to enable orbital expansion (10:19).

• A Potential New Industry

  • If feasible, this could spark a boom in spacecraft launches and a new sector around space-based AI computation.
  • However, the endeavor is “unproven with both engineering and technical and immense cost hurdles” (11:30).

Notable Quotes

• On the promise and skepticism of the orbital race:

  “It’s just, you have to come back to time and again that this is just an idea at this point. This is unproven with both engineering and technical and immense cost hurdles.” —Micah Madenberg [11:30]

Memorable Moments

• Host Katie Dayton’s summary of the “2025 story trifecta”—Musk, Bezos, and AI—captures the episode’s zeitgeist (06:35).

---

Notable Quotes with Timestamps

• Ellen Gammerman:

  “She’s a little bit sassy, but she’s also a little bit weirdly self-aware. Like she calls herself the end of civilization because so many people have been so upset about her being created.” [01:47]

• Ellen Gammerman:

  “It’s tapping into these fears that everyone there is replaceable, including now actors… this could start seeping into the creative arena.” [03:35]

• Ellen Gammerman (via Van der Velden):

  “AI can build worlds in ways that traditional filmmaking can’t… there’s a kind of a creative renaissance.” [04:00]

• Micah Madenberg:

  “The pull factor is literally straight out of science fiction… we’re at the point where launch costs and satellite component production costs maybe are low enough to make that science fiction dream a reality.” [07:48]

• Micah Madenberg:

  “It’s just, you have to come back to time and again that this is just an idea at this point. This is unproven with both engineering and technical and immense cost hurdles.” [11:30]

---

Timestamps of Important Segments

• [00:19] — Introduction to Tilly Norwood, the AI actress
• [01:41] — Building Tilly’s personality and her “sassy” traits
• [02:29] — Background of her creator, Aline van der Velden
• [03:18] — Industry backlash and James Cameron’s criticism
• [04:00] — Van der Velden’s vision for AI film as an art form
• [05:03] — Secret Hollywood interest; NDAs and the future of AI movies
• [06:35] — Shift to the orbital AI data center “space race”
• [07:16] — Rationale for moving data centers to space
• [08:24] — Feasibility: design based on massive satellite networks
• [09:01] — Practical and financial obstacles
• [10:19] — Current investments, major players, and plans for experiments
• [11:30] — The nascent state of the orbital data center industry

---

Tone & Style

The episode is investigative and future-focused, mixing journalistic skepticism with fascination for radical new frontiers. The host and guests blend accessible explanations with industry insights, all in the concise, authoritative tone characteristic of WSJ reporting.

---

Summary

This episode explores the radical ambitions reshaping both Hollywood and global tech infrastructure. The arrival of AI-generated actors like Tilly Norwood provokes existential anxiety in creative industries, while the race to run AI from space is rife with both promise and skepticism—underscoring the disruptive, still uncharted world unfolding at the intersection of AI and human enterprise.

Transcript

[A]

Every organization, government or commercial, is being asked to deliver more with less.

[B]

Technology is undeniably the key to making that possible.

[C]

We invite you to explore the insights in the Tech Guide, available [e-mail].

[A]

Welcome to Tech News briefing. It's Friday, December 12th. I'm Katie Dayton for the Wall Street Journal. Have you heard about Tilly Norwood? The actress primed to take Hollywood by storm? She's beautiful, she's adapt, and she's not real. We investigate why a former actress spent months generating the perfect AI counterpart and became the talk of Hollywood in the process. Then we're exploring the new frontier of orbital data centers. Elon Musk and Jeff Bezos are leading the push to take the gigantic infrastructure required to power AI and launch it into space. Stay tuned to find out how what sounds like a science fiction dream may become a reality. But first, Tilly Norwood is a brown haired, brown eyed, 24 year old actress who would have big dreams if she was sentient. She's an AI generated creation that took six months to build and garnered more than a few negative headlines when she was revealed to the world as the first AI movie star. But not much is known about Tilly's creator or what her career prospects might look like from here. WSJ reporter Ellen Gammerman has been working to find out. So Ellen, tell us about Tilly. Does she have any discernible personality traits aside from being astoundingly beautiful?

[C]

Well, that's what's happening right now. Her creator, Eline van der Velden is right now building her brain, which means building out her personality and her behavior and eventually even how she can respond one on one with fans. What we've seen right now from promotional videos and little snippets is that she's a little bit sassy, but she's also a little bit weirdly self. Aware. Like she calls herself the end of civilization because so many people have been so upset about her being created.

[A]

Bit of a sarcastic quality to her then. So you mentioned her creator, Aline Van de Velden. Can you tell us a little bit about who she is and why she would go to all this trouble to make this character?

[C]

She lives in London. She had been working on AI production with her own studio and it had been mostly hybrid projects. But she was getting into full AI films, completely computer made and she started to notice AI influencers on her feed and she thought, well, AI movie stars are just the next thing that's going to happen, so why not create one right now? And the reason she felt qualified is because she's a former actress and when she wasn't acting, she was making little comic videos for YouTube and BBC3 where she would create these characters. So she felt like she already knew how to build the ideal movie starlet.

[A]

And it's fair to say that the reaction to Tilly in Hollywood so far has not been positive. You have the legendary director James Cameron describing her as horrifying at one point. What kind of fears over the future of Hollywood is this all tapping into?

[C]

It's tapping into these fears that everyone there is replaceable, including now actors, and that already they're seeing jobs go to AI that used to go to people, but in the technical arena. So the real fear is that this could start seeping into the creative arena.

[A]

And how has Aline responded to those criticisms?

[C]

Look, she is a former actress. She feels kinship with other actors. She does not want to replace them. What she sees is this being its own art form. AI can build worlds in ways that traditional filmmaking can't. And she imagines a world where there's a kind of a creative renaissance where these filmmakers who specialize in AI are building out their own characters and scenarios that are completely separate from and not in competition with traditional filmmaking. There's a lot of skepticism about that from people who make their living making films the traditional way.

[A]

So the plan for Tilly is to become a Hollywood star. What is she at this point in time? Obviously we haven't quite yet seen AI characters and actors appearing in films yet. So is she auditioning for anything? Is she actually acting at this point? How does all of that work?

[C]

It's such a good question. It's kind of happening already. According to Van der Velden, who says she's signed more than 60 non disclosure agreements with various Hollywood players, different producers and people who are very interested in making either partly AI or fully AI films. And everyone is so freaked out about it that no one wants their name to be known. So all of her projects are very hush hush right now. But I do wonder what they will look like in the final form and if this is really going to turn into an actual AI movie start. We still don't know.

[A]

That was WSJ reporter Ellen Gammerman. Would you watch a movie featuring an AI actor? If you're a listener on Spotify, be sure to let us know in this episode's poll or leave us a comment. Coming up. President John F. Kennedy once called Spider space the new frontier. Now space itself is becoming a new frontier for AI data centers. That's after the Break.

[C]

For government and business leaders, AI isn't the future.

[A]

It's the force reshaping today.

[C]

The Tech Guide helps organizations move from exploration to execution. Check out the Tech Guide, available [e-mail].

[A]

A new space race is on, and this time it's all about data. Space companies backed by tech billionaires on Earth are hoping to move AI data centers into orbit. Jeff Bezos Blue Origin has been working for more than a year on the technology needed to make it happen, While Elon Musk's SpaceX plans to use an upgraded version of its Starlink satellites to host AI computing payloads. WSJ space reporter Micah Madenberg has the details. So Micah, your article has all the components of a 2025 story. You've got Musk, Bezos, AI. Can you start by explaining why some say we need data centers in space?

[B]

Yeah, there's a push and a pull. The push is the data center construction rush here on the ground is rife with complexity and headaches, supply chain delays, construction issues, and energy and power. Where do you get the electricity to run data centers? The pull factor is literally straight out of science fiction. People have been talking about harnessing the immense power of the sun in space for applications back on the ground for decades. And when you talk to folks in space and AI or at the intersection of both of those, their argument is we're at the point where launch costs and satellite sort of component production costs maybe are low enough to make that science fiction dream a reality.

[A]

These data centers on Earth are absolutely huge, which begs the question of how you'd ever get something that size into space. Do these companies that are interested in this have a strategic idea of how they're going to make that work?

[B]

So right in space, as people are trying to get their heads around the design here, think of smaller, perhaps thousands upon thousands of satellites working in sort of a cluster, working as like a network. There are some technical and sort of physics based reasons why actually having smaller devices, but many of them that are linked up with each other and transmitting data back to the ground make more sense than doing these huge devices.

[A]

And what kind of costs, monetary and otherwise, are we talking about here?

[B]

There's not a lot of precedent for say getting 10,000 satellites operating as a cluster to provide gigawatt scale AI computing services in space. I mean, SpaceX has around 9,000 satellites for its Starlink satellite business and that's the largest satellite fleet deployed in history. So just to build the number of satellites, that's very difficult and costly. And then to Launch them over and over and over. That's also a long process, difficult and costly. And so even with lower launch costs, with better components, electronics tech for satellites, you're talking about costs that pile up really, really quickly. And that leads a lot of people to sort of look at this idea and say, it just, it ain't going to happen. And the dynamics on the ground will also continue to change. Maybe new power sources come online, supply chain constraints ease, and so on and so forth. So that's part of the debate that's happening right now.

[A]

So a huge effort required. How much are companies putting into making this happen? How seriously are they taking this?

[B]

Google is working with this company, a satellite operator and builder called Planet Labs, and they have described their efforts here as an experiment. They're going to send a couple of AI compute satellites to orbit in early 2027, test out the systems and try to get data to move forward. Musk is making significant pronouncements about what may be possible. Over at SpaceX, Jeff Bezos's space company, Blue Origin, has had a team working on orbital data centers for more than a year. They're very serious about this, as I understand it. And then there's a bunch of other companies that are also trying to, like, figure out how they would get involved here. Sam Altman, the CEO of OpenAI, has investigated trying to find a way to map OpenAI onto a rocket company because again, if you're going to send potentially thousands of AI compute satellites to orbit, you need a way to get them there.

[A]

So many players at the top of this game. I wonder if this has the potential to create a whole new industry of launching and running these space data centers.

[B]

For, like, rocket launchers. The prospect is very exciting because rockets are expensive, difficult, complicated, they're time consuming to build and, and then when you build one, you've got to find payloads to like, send to space. But it's just, you have to come back to time and again that this is just an idea at this point. This is unproven with both engineering and technical and immense cost hurdles. So we need to see how this stuff starts to prove out beyond the discussions and presentations and those sorts of things.

[A]

That was WSJ reporter Micah Madenberg. And that's it for Tech News Briefing. If you're a listener on Spotify, be sure to take this episode's poll or leave us a comment. Today's show was produced by Julie Chang. I'm your host, Katie Dayton. Jessica Fenton and Michael Lavalle wrote our theme music. Our supervising producer is Katie Ferguson. Jessica Fenton is our technical manager. Our development producer is Aisha Al Muslim. Chris Sinsley is the deputy editor. And Falana Paterson is the Wall Street Journal's head of news audio. We'll be back later this morning with TNB Tech Minute. Thanks for listening. Every organization, government or commercial, is being asked to deliver more with less.

[B]

Technology is undeniably the key to making that possible.

[C]

We invite you to explore the insights in the Tech Guide, available [e-mail].