Peter Diamandis
Episode overview

The $10B Satellite Empire Putting AI in Orbit, Why Chips Beat Rockets & China's #1 Open Model | EP #266

Peter Diamandis · 26 June 2026 · 5 Egleze moments
The $10B Satellite Empire Putting AI in Orbit, Why Chips Beat Rockets & China's #1 Open Model | EP #266 ▶

What happened in this episode?

In this episode of Moonshots, host Peter Diamandis is joined by co-hosts Salim Ismail, Alex Guzey, and Dave Blunden, along with special guest Will Marshall, CEO of Planet, the world's largest Earth-observing satellite fleet. The conversation centers on Planet's push to build what Marshall calls 'large Earth models'—AI systems trained on 150 petabytes of daily satellite imagery spanning a decade, enabling real-world queries about agriculture, disasters, and geopolitics. Marshall revealed a critical insight: Google's TPU energy efficiency per inference, not launch costs, will determine the winner of the orbital AI data center race, as compute efficiency drives spacecraft mass and cooling requirements. The episode covered Eric Schmidt's acquisition and assumption of CEO role at Relativity Space, positioning it as an alternative to SpaceX's launch monopoly for space-based compute. Major talent defections from Google DeepMind—including Transformer inventor Noam Shazier to OpenAI and Nobel laureate John Jumper to Anthropic—sparked debate over whether Google has fallen behind the frontier. The hosts analyzed China's GLM 5.2 open-weight model, which matches top Western labs on key benchmarks by burning twice the reasoning tokens at half the cost, challenging assumptions about permanent Chinese lag. Argentine President Javier Milei's proposal for full AI legal personhood drew sharp rebuttals from Yuval Harari, framing a global debate on machine accountability. Throughout, the conversation returned to existential questions: alignment, recursive self-improvement, the Fermi paradox, and whether transparency from orbital sensing can create 'planetary wisdom' to navigate humanity's most consequential technological transition.

What are the key points?

Which topics are covered?

How is this page sourced?

Programme
Peter Diamandis
Published
26 June 2026
Original episode
Open the original source ↗

Egleze verifies the speaker, source and timestamp for each linked moment. Unless a moment is explicitly marked independently verified, statements remain attributed claims by the named speaker rather than established facts.

Watch original episode More from Peter Diamandis

5 moments from this episode

Source-linked · editorially selected
01
AI & Tech

Eric Schmidt Purchases Relativity Space to Compete with SpaceX Monopoly

at 1:07:57 · Attributed speaker claim
Eric is the CEO now of Relativity Space, which blew my mind when he took that role. And they just announced they have gotten a mission from NASA called EULIS.

Former Google CEO Eric Schmidt acquired struggling rocket company Relativity Space after its financing collapsed and personally took over as CEO. The company secured NASA's EULIS Mars mission shortly after. Schmidt, an early investor in Planet, appears to be positioning Relativity as an alternative launch provider for orbital AI data centers, directly challenging SpaceX's emerging monopoly on space infrastructure.

02
AI & Tech

Planet CEO Reveals TPU Efficiency Determines Winner of Orbital AI Race

at 1:17:54 · Attributed speaker claim
Everyone apart from SpaceX has to pay the SpaceX launch tax right now. Everyone apart from NVIDIA and Google has to pay the NVIDIA tax. And which tax is more important? Near-term is the launch, but longer-term is the compute.

Will Marshall, CEO of Planet, revealed on Moonshots that Google's TPU energy efficiency per inference—not launch costs—will determine dominance in orbital AI data centers. He argued that compute efficiency drives radiator mass and spacecraft cost, making Google's TPU advantage potentially more significant than SpaceX's launch monopoly. This insight reframes the space AI race around chip design rather than rockets.

03
AI & Tech

Google Loses Two Nobel-Level AI Scientists to OpenAI and Anthropic

at 1:25:37 · Attributed speaker claim
Noam Shazier, he was the lead author in the Transformer paper, which is the T in GPT. He's unfortunately leaving Google. For OpenAI. And get this, this is the second time he's left Google.

Noam Shazier, inventor of the Transformer architecture underlying all modern AI, departed Google DeepMind for OpenAI just two years after Google bought his company Character.AI for $2.7 billion to bring him back. Nobel laureate John Jumper, co-creator of AlphaFold, also left Google DeepMind for Anthropic. The dual defections suggest Google has fallen behind the frontier as top researchers seek raw access to pre-trained models without guardrails.

04
AI & Tech

Argentina Proposes Full Legal Personhood for AI Corporations, Yuval Harari Objects

at 1:44:41 · Attributed speaker claim
Milei wrote a letter to Yuval Harari saying he proposes that AI should be able to incorporate sign contracts, hire people, and sue with no humans in the loop. Legal persons allow for accountability.

Argentine President Javier Milei proposed creating a legal category allowing AI to incorporate, sign contracts, hire employees, and sue without human oversight, arguing it enables accountability through asset seizure. Yuval Harari published a rebuttal warning this creates a non-human shield where citizens could be ruled by entities lacking moral accountability. The debate represents the first major governmental push for AI personhood, with Argentina positioning itself as the global home for unregulated AI development.

05
AI & Tech

China's GLM 5.2 Model Matches Top US Frontier AI Labs

at 1:59:17 · Attributed speaker claim
This level of performance in an open-weight model is absolutely shocking. You can burn tokens to get more intelligence, and the Chinese have figured out how to do it.

Chinese lab Zhipao AI released GLM 5.2, a 753-billion-parameter open-weight model that matches or exceeds OpenAI and Anthropic on key benchmarks, especially reasoning tasks. The model achieves similar performance to Opus 4.8 at roughly half the cost by using double the reasoning tokens. This challenges the assumption that Chinese models lag Western labs by 6-8 months and demonstrates frontier intelligence can no longer be monopolized.