Thinking On Paper

Can humanoids dance? Or will billions of Tesla robots choose to forgo such technological frivolity?Philip Johnston is CEO of Starcloud. They build data centers in space. Their first satellite just launched on SpaceX Falcon 9. You can track it orbiting Earth right now.This short covers humanoid robots, data centers in orbit, and whether the future includes dancing machines.We talk about:- Why Philip predicts 5 billion humanoids by 2035- What humanoid robots will actually do (not dance—work)- How space-based data centers solve Earth's power crisis- Why orbit is better for computing than ground (cooling, energy, latency for some tasks)- The Starcloud satellite currently in orbit (track it yourself)- Whether robots need to be humanoid at all (or if we're copying ourselves for ego)Philip's thesis: Humanoids scale faster than anyone expects. Not because they're better than specialized robots—but because they navigate human infrastructure without redesigning the world.The question: Do we need 5 billion humanoids? Or do we just think we do because they look like us?This is a short from the full conversation. Listen to the complete episode for more on space infrastructure, orbital manufacturing, and the future Philip's building.---Guest: Philip Johnston, CEO, StarcloudTopics: Humanoid robots, space data centers, satellites, SpaceX, infrastructure, robotics, orbital computingStatus: Starcloud satellite currently in orbitFormat: Short episode | Full version available--Other ways to connect with us:⁠Listen to every podcast⁠Follow us on ⁠Instagram⁠Follow us on ⁠X⁠Follow Mark on ⁠LinkedIn⁠Follow Jeremy on ⁠LinkedIn⁠Read our ⁠Substack⁠Email: hello@thinkingonpaper.xyz

AI agents can read feeds, make decisions, coordinate with other agents, and speak on your behalf—without you in the loop.Andrew Hill explains what agents actually are, why every company is racing to build them, and how close we are to personal agents that manage schedules, explain our thinking, and negotiate with other people's agents.We talk about:- What an AI agent actually is (beyond chatbots)- Why agents coordinate with each other (multi-agent systems)- How personal agents could represent you online- What happens when your agent negotiates with someone else's agent- Why people already share intimate details with AI (and what that means)- The hard question: Could AI become better relationship partners than humans?The shift that's already happening: People tell AI things they won't tell friends. They trust agents with calendars, emails, thoughts. The AI knows them better than anyone.So if agents represent us online—if they speak for us, decide for us, negotiate for us—who are we really talking to anymore?This gets into trust, privacy, and what changes when the agentic web replaces direct human interaction.If your agent knows you better than your partner does, what does that make it?---Guest: Andrew HillTopics: AI agents, agentic web, multi-agent systems, personal AI, trust, privacy, human-AI relationships, coordination--Other ways to connect with us:⁠Listen to every podcast⁠Follow us on ⁠Instagram⁠Follow us on ⁠X⁠Follow Mark on ⁠LinkedIn⁠Follow Jeremy on ⁠LinkedIn⁠Read our ⁠Substack⁠Email: hello@thinkingonpaper.xyz

To survive in space, you don't just need engineers. You need a musician. Preferably a guitarist.Jeremy asks physicist Danny Andreev (CEO, Sunburn Schematics): Could my 1969 Fender Vibrolux amp work in space?Answer: Yes. Analog gear shrugs off radiation.What starts as electrical engineering turns into human psychology and Mars survival.We talk about:- Why Jeremy's vintage guitar amp would work on the moon (analog circuits resist radiation)- What modifications it would need (thermal management, vacuum considerations)- How digital devices fail in space while analog survives- Why submarines and Arctic research stations need musicians (group cohesion studies)- How having a guitarist changes crew survival in isolated environments- Why Mars missions need musicians, comedians, and risk-takersThe research: Studies on submarines and Antarctic bases show musicians are critical for group survival. Not nice-to-have. Critical.Music affects morale, bonding, and psychological resilience in ways nothing else does.Elon, if you're listening: You need guitarists on those Mars ships. Not for fun—for survival.This isn't a gear review. It's about culture, isolation, and what humans actually need when they're far from home.Rock on.---Guest: Danny Andreev, Physicist, CEO Sunburn SchematicsTopics: Space electronics, Mars missions, musicians, isolation, group psychology, analog vs digital, radiationFun fact: Vintage amps work in space--Other ways to connect with us:⁠Listen to every podcast⁠Follow us on ⁠Instagram⁠Follow us on ⁠X⁠Follow Mark on ⁠LinkedIn⁠Follow Jeremy on ⁠LinkedIn⁠Read our ⁠Substack⁠Email: hello@thinkingonpaper.xyz

Curious Minds Learn about THE REAL IMPACT of technology 👇 Thinking on Paper goes holistic. Your learning goes ballistic.Other ways to connect with us:⁠Listen to every podcast⁠Follow us on ⁠Instagram⁠Follow us on ⁠X⁠Follow Mark on ⁠LinkedIn⁠Follow Jeremy on ⁠LinkedIn⁠Read our ⁠Substack⁠Email: hello@thinkingonpaper.xyzAI, Quantum computing, space manufacturing, robotics and Web3! From the CEOS and Silicon Valley Founders spending millions and billions making them useful.Or destroying the planet for their Egos. Take your curiosity, push it to its limits and see what technology can really do.Our mission is to help ONE MILLION curious minds ditch their Twitter and LinkedIn feeds and connect the dots for themselves. Each week, hosts Mark and Jeremy take you inside IBM, NASA, Coinbase, D-Wave, and more. They focus on how systems work, what they cost, who benefits, and the impact on work, policy, culture, and family. There's a Book Club too. Because the oldest tech is still the best. Stop scrolling and subscribe.

AI Ethics is a mirage straight from a Kafka novel. Questions of justice, principles and the rule of law are incompatible with machine learning. Machine learning is statistical analysis of data that outputs responses human beings are likely to find attractive, not true or ethical. That is not a good way to design ethics.Carissa Veliz joins Makr & Jeremy to Think On Paper. She outlines how AI depends on surveillance and statistical pattern-matching that can’t meet the basic standards of a democracy: clear rules and the ability to appeal a decision.AI thinking clashes with the foundations of a liberal democracy: public rules, transparency, and the right to challenge decisions that shape your life.We cover:😀 Why machine-learning decisions are opaque😀 Why that conflicts with the rule of law😀 How surveillance sits beneath modern AI systemsPlease enjoy the show.Cheers, Mark & JeremyOther ways to connect with us:⁠Listen to every podcast⁠Follow us on ⁠Instagram⁠Follow us on ⁠X⁠Follow Mark on ⁠LinkedIn⁠Follow Jeremy on ⁠LinkedIn⁠Read our ⁠Substack⁠Email: hello@thinkingonpaper.xyz

Exxon and Chevron are using Microsoft AI to extract more oil. Faster. Cheaper. The goal: every last drop.Holly and Will Alpine (Enabled Emissions) paint a stark picture with Microsoft's own numbers:Exxon deal: 50,000 barrels/day = 6.4 million tonnes CO₂/yearChevron deal: 400,000 barrels/day = 51 million tonnes CO₂/yearMicrosoft's entire FY23 footprint: 17 million tonnes.Carbon removal booked over 15 years: 5 million tonnes.Those two deals alone dwarf both numbers.In Saudi Arabia, Aramco's CEO says AI kept production costs at $3/barrel for two decades. AI makes fossil fuels competitive. It weakens clean energy economics.AI touches every stage of fossil fuel: exploration, drilling, extraction, refining, distribution.We talk about:- How AI cuts production costs 10% while boosting reserves 5%- Why US oil production tripled since 2007 (AI is a major factor)- Projects that took 18 months now take 2 weeks- How aging oil fields stay profitable decades longer- Why Microsoft's "energy principles" are meaningless (they ignore Scope 3)- What guardrails could look like (restrict exploration/extraction, allow safety/methane reduction)Holly and Will left Microsoft to start Enabled Emissions. They're not asking companies to break contracts. They're asking for reasonable guardrails.The lie you're fed: AI will solve climate change.The truth: AI is accelerating fossil fuel extraction.It's ugly. It's real. Learn more.---Guests: Holly & Will Alpine, Founders Enabled Emissions (Former Microsoft)Topics: AI, climate, oil, emissions, Microsoft, Exxon, Chevron, Aramco, enabled emissions--Other ways to connect with us:⁠Listen to every podcast⁠Follow us on ⁠Instagram⁠Follow us on ⁠X⁠Follow Mark on ⁠LinkedIn⁠Follow Jeremy on ⁠LinkedIn⁠Read our ⁠Substack⁠Email: hello@thinkingonpaper.xyz

What is a spin qubit?Brandon Severin (CEO, Conductor Quantum) explains: Take one electron. Put it in a magnetic field. It acts like a tiny compass needle with two orientations—spin-up and spin-down.Those are your 0 and 1.Isolate that electron on a gated silicon device. Hit it with precise pulses. You can flip, hold, and combine those states (superposition). That's quantum computing.The advantage: Spin qubits use the same fabrication tech as classical transistors.Modern NVIDIA and Apple chips have tens of billions of transistors. The same infrastructure could eventually produce comparable numbers of spin qubits. Each qubit is just one electron you can address and control.We talk about:- How spin-up and spin-down create quantum states- Why silicon fabrication gives spin qubits massive scaling potential- How you flip electron spins with precise pulses- What superposition means at the electron level- Why spin qubits could scale faster than superconducting or trapped ion systems- The challenge: controlling billions of individual electrons with precisionThe promise: If we can print 50 billion transistors on a chip, we could eventually print 50 billion qubits.That's the quantum leap spin qubits are betting on.---Guest: Brandon Severin, CEO Conductor QuantumTopics: Spin qubits, quantum computing, electrons, silicon fabrication, superposition, scalingWatch on YouTube: https://www.youtube.com/watch?v=9LeN3VBvG0o&t=1sCheers,Mark & Jeremy--Other ways to connect with us:⁠Listen to every podcast⁠Follow us on ⁠Instagram⁠Follow us on ⁠X⁠Follow Mark on ⁠LinkedIn⁠Follow Jeremy on ⁠LinkedIn⁠Read our ⁠Substack⁠Email: hello@thinkingonpaper.xyz

Mind Blowing to the banks, but Stablecoins already move more volume than Visa and Mastercard combined! There's a financial revolution in the air, and this time, you're invited.Robby Yung, CEO of Animoca Brands, shows how people move dollars across borders in minutes with near-flat fees, from market traders in Nigeria to institutions shifting tens or hundreds of millions. This is a short from our full length deep dive into web3, the decentralized internet, DOAs, AI and what Animoca has in store for the coming year.Subscribe to Thinking on Paper for the full conversation.--Other ways to connect with us:⁠Listen to every podcast⁠Follow us on ⁠Instagram⁠Follow us on ⁠X⁠Follow Mark on ⁠LinkedIn⁠Follow Jeremy on ⁠LinkedIn⁠Read our ⁠Substack⁠Email: hello@thinkingonpaper.xyzWatch On YouTube: https://youtu.be/O_Iy1jYTRz8

China built its space station Tiangong in three years after being excluded from the ISS. It landed on the Moon twice—2020 and 2024—returning samples from high helium-3 areas. Now the race is for resources and rules.Glen Martin (aerospace designer, ISS contributor) explains how China's space program connects: high-speed rail expertise at home translates to orbital infrastructure. Grid power systems scale to space stations. Industrial coordination enables lunar missions.The timeline: 2029 crewed lunar landing planned for the 80th anniversary of the 1949 revolution. They're on track.What they've already done:- Built Tiangong space station (operational, continuously crewed)- Landed on lunar far side using relay satellites and nuclear-powered rover- Collected samples from helium-3 rich regions (twice)- Demonstrated industrial-speed space development (3 years for Tiangong vs ISS decades)Why the lunar south pole matters: "Rim of eternal light" has 24-hour sunlight plus water ice in permanently shadowed craters. Perfect for long-term habitation.The rules fight:- UN Space Resources Treaty draft due 2027 (two years before China's planned landing)- US domestic space laws say "just go mine it" (no international permission needed)- Deep sea mining comparison: UN trying to regulate, countries acting unilaterally- Wild West vs regulated approach—which wins?Who decides what happens on the Moon? Right now, nobody. That's the problem—and the opportunity.China wasn't invited to play ISS. So they built their own game.---Guest: Glen Martin, Aerospace Designer | ISS Contributor, Extraterrestrial Mining Company CEOTopics: China space program, Tiangong, Moon landing, helium-3, space law, UN treaty, lunar resourcesFormat: Short episode--📺Watch on YouTube--TIMESTAMPS(00:00) Chinese Infrastructure(00:47) Bringing Russia to ISS(01:21) We Blocked The Chinese(01:41) Tiangong & 2029 Moon Landings(02:22) The Global Politics Of Space(03:36) The Lunar South Pole(04:07) The United Nations(04:41) The Moon Wild West--Other ways to connect with us:⁠Listen to every podcast⁠Follow us on ⁠Instagram⁠Follow us on ⁠X⁠Follow Mark on ⁠LinkedIn⁠Follow Jeremy on ⁠LinkedIn⁠Read our ⁠Substack⁠Email: hello@thinkingonpaper.xyz

You're listening to AI-generated music and don't realize it. The musicians whose work trained those models know. They check their empty bank accounts daily.99,000 new songs upload to streaming platforms every day. One in five are AI-generated (Deezer). You wouldn't play a single one at your funeral.59% of musicians use AI in some aspect of their music (Ditto Music). The question: How do real musicians get paid when AI uses their work?We break down Water & Music's research on AI music attribution. Cherie Hu, Yung Spielburg, and Alexander Flores investigated what's at stake and which companies are trying to solve it.The problems:- Session musicians, producers, songwriters—how do they get paid when AI uses their beats?- Record labels can't track which training data influenced which outputs- Proving a musician's input in a model's output is nearly impossible- Copyright law wasn't built for this- Most AI music companies scraped without permission or paymentWe talk about:- How attribution technology could work (vector matching, contribution tracking)- Which companies are building payment systems- Why Suno and Udio's approach creates legal chaos- How ethical companies like Overtune license training data and split royalties- Whether streaming platforms can detect AI-generated music- What happens when the Beatles' catalog trains a modelThe stakes: If attribution fails, AI becomes theft at industrial scale. Music becomes disposable content optimized for algorithms, not humans.Based on Water & Music's research—the best reporting on AI music economics.Share with a music lover.---Research: Water & Music (Cherie Hu, Yung Spielburg, Alexander Flores)Topics: AI music, copyright, attribution, streaming, royalties, training data–(00:00) The Intersection of Music and AI(03:26) Understanding Music Attribution(03:51) Sonic Characteristics and AI Influence(06:39) The Complexity of AI Music Generation(07:36) The Value Equation in AI Music Creation(08:08) Understanding Influence Functions in Music AI(09:44) Challenges of Attribution in AI-Generated Music(11:38) Exploring Embeddings and Their Role in Music AI(14:17) Watermarking and Its Limitations in Music Attribution(15:30) Synthetic Data and Its Implications for Music AI(17:48) Innovative Solutions for Music Rights Attribution(18:01) Distinguishing Compositional vs. Recording Contributions(19:59) The Impact of AI on the Music Industry's Inequities(23:03) Trust and Technology in Music AttributionOther ways to connect with us:⁠Listen to every podcast⁠Follow us on ⁠Instagram⁠Follow us on ⁠X⁠Follow Mark on ⁠LinkedIn⁠Follow Jeremy on ⁠LinkedIn⁠Read our ⁠Substack⁠Email: hello@thinkingonpaper.xyz