World models: LLMs are not enough

The web is turning agentic. And that changes everything from shopping to search to SEO.In this episode of TechFirst, John Koetsier sits down with Dave Anderson (VP at ContentSquare + host of the “Tech Seeking Human” podcast) to unpack what happens when browsers and AI assistants don’t just answer … they do stuff. For you. On your behalf.From Atlas and agentic browsing to the growing backlash from retailers (hello, Amazon vs Perplexity), we explore who benefits, who loses, and what the internet becomes when agents are the default user.You’ll hear why retailers are nervous (security, margins, coupon hunting), why agent-first experiences might create “headless” retailers (like ghost kitchens, but for ecommerce), and why search is shifting from SEO to AI visibility. Plus: real talk about trusting agents with your credit card, hallucinations, and what it means if your agent can look indistinguishable from you.GuestDave Anderson — VP, ContentSquarehttps://contentsquare.comPodcast: Tech Seeking Humanhttps://www.techseekinghuman.aiLinks & subscribeSubscribe for more conversations on tech, AI, and what’s next: https://techfirst.substack.comTranscripts always available herehttps://johnkoetsier.com00:00 Agentic web: what changes when browsers “do stuff”00:59 Meet Dave Anderson (VP + podcast host)01:31 30,000 feet: why “agents” suddenly matter03:48 The agent future John wanted 10 years ago04:21 Why Amazon doesn’t want your agent shopping on Amazon05:07 Ticketmaster, bots, and the security nightmare06:26 Siri’s original promise vs today’s reality08:31 Are agents just bots… or something different?10:04 Retail fears: coupon hunting, margins, returns chaos11:21 Can you trust an agent with your credit card?11:59 Why retailers want their own agents (and control)13:14 Amazon’s agent works… but is it the whole internet?14:19 Ghost kitchens for retail: “headless” agent-first brands15:17 Hugo Boss jacket test: agents vs manual search16:40 Agents should talk to your finance agent17:14 Kids + deepfakes: what even looks real anymore?18:04 Is this corrosive to apps… or the web?19:10 Online identity, anonymity, and agent verification20:28 Two futures: human-first brands vs agent-first retail21:19 Agentic browsers on your device: can they “look like you”?22:51 Baseball vs golf: the best analogy for search now24:44 Instant shopping problem: returns + missing “services layer”26:10 AI weirdness: wrong names, wrong locations, shifting behavior27:37 Agents beyond shopping: support is the sleeper win29:49 Inventing the future: who adopts agents and who won’t31:13 Will people get tired of AI and crave humans again?31:45 Serendipity vs optimization: the restaurant debate32:36 Wrap: nobody solved agents… but the shift is real

AI has mastered language, sort of. But the real world is way messier.In this episode of TechFirst, John Koetsier sits down with Kirin Sinha, founder and CEO of Illumix, to explore what comes after large language models: world models, spatial intelligence, and physical AI.They unpack why LLMs alone won’t get us to human-level intelligence, what it actually takes for machines to understand physical space, and how technologies born in augmented reality are now powering robotics, wearables, and real-world AI systems.This conversation goes deep on: • What “world models” really are — and why everyone from Fei-Fei Li to Jeff Bezos is betting on them • Why continuous video and outward-facing cameras are so hard for AI • The perception stack behind robots and smart glasses • Edge vs cloud compute — and why latency and privacy matter more than ever • How AR laid the groundwork for the next generation of physical intelligenceIf you’re building or betting on robotics, smart wearables, AR, or physical AI, this episode explains the infrastructure shift that’s already underway.GuestKirin SinhaFounder & CEO, Illumixhttps://www.illumix.com👉 Subscribe for more deep conversations on technology, AI, and the future:https://techfirst.substack.com00:00 Raising the Bar on “Smart” Devices01:07 Meet Kirin, Founder & CEO of Illumix01:21 What Is a World Model — and Why It Matters02:23 Why LLMs Alone Won’t Lead to AGI03:46 From AR & the Metaverse to Physical AI05:18 AR vs VR vs the Metaverse — Different Problems, Different Futures06:32 Spatial Perception, Scene Understanding, and Contextual Intelligence07:39 Why Continuous Video Is So Hard for Machines08:39 The Camera Flip: From Selfie AI to World-Facing AI09:58 Why Cameras Beat LiDAR for Wearables and Robots10:27 Inside the Perception Stack11:20 Edge vs Cloud Compute in Physical AI12:37 Why On-Device Intelligence Matters for UX13:52 SLMs, Efficiency, and the Limits of “Bigger Is Better”15:11 Knowing What to Run — and When16:06 Intent, Memory, and Real-Time AI Decisions17:32 Physical Intelligence vs Digital Intelligence18:39 Memory Palaces, Spatial Brains, and Human AI19:39 Do We Need New Chips for Humanoid Robots?20:26 How Chip Architectures Will Evolve for Physical AI21:47 Privacy, On-Device Processing, and Trust22:48 Final Thoughts on the Future of World-Aware AI

Quantum computers usually mean massive machines, cryogenic temperatures, and isolated data centers. But what if quantum computing could run at room temperature, fit inside a server rack — or even a satellite?In this episode of TechFirst, host John Koetsier sits down with Marcus Doherty, Chief Science Officer of Quantum Brilliance, to explore how diamond-based quantum computers work — and why they could unlock scalable, edge-deployed quantum systems.Marcus explains how nitrogen-vacancy (NV) centers in diamond act like atomic-scale qubits, enabling long coherence times without extreme cooling. We dive into quantum sensing, quantum machine learning, and why diamond fabrication — including the world’s first commercial quantum diamond foundry — could be the key to manufacturing quantum hardware at scale.You’ll also hear how diamond quantum systems are already being deployed in data centers, how they could operate in vehicles and satellites, and what the realistic roadmap looks like for logical qubits and real-world impact over the next decade.Topics include: • Why diamonds are uniquely suited for quantum computing • How NV centers work at room temperature • Quantum sensing vs. quantum computing • Manufacturing challenges and timelines • Quantum computing at the edge (satellites, vehicles, sensors) • The future of hybrid classical-quantum systems⸻🎙 GuestMarcus DohertyChief Science Officer, Quantum BrillianceProfessor of Quantum PhysicsArmy Reserve Officer🌐 https://quantumbrilliance.com⸻👉 Subscribe for more deep dives into the future of technology:https://techfirst.substack.com⸻00:00 Diamonds and the next wave of quantum computing01:20 Why diamond qubits work at room temperature03:20 NV centers explained: defects that behave like atoms05:05 How diamonds replace massive quantum isolation systems06:40 Building the world’s first quantum diamond foundry08:30 Defect-free diamonds, isotopes, and qubit engineering10:15 Quantum sensing vs. quantum computing with diamonds12:40 From desktop quantum systems to millions of qubits14:25 Roadmap: logical qubits, timelines, and scale16:10 Quantum computers at the edge: vehicles and satellites18:10 Quantum machine learning and real-world deployments19:50 The long game: why diamond quantum computing scales

Will AI kill your job?What happens to your job as AI gets smarter and companies keep laying people off even while profits rise? Will you still have a job? Will the job you have change beyond recognition?Scary questions, no?In this episode of TechFirst, host John Koetsier sits down with Nikki Barua, co-founder of Footwork and longtime founder, executive, and resiliency expert, to unpack what work really looks like in the age of AI.Layoffs are no longer just about economic downturns. Companies are growing, innovating, and still cutting staff, often because AI is enabling more output with less capacity. So what does that mean for you?Nikki argues the future doesn’t belong to those who simply “learn AI tools,” but to agentic humans: people who lead with uniquely human strengths and use AI to amplify their impact. This conversation explores:• Why today’s layoffs are different from past cycles• How AI is compressing jobs before creating new ones• What it means to move from doing work to directing outcomes• Why identity, curiosity, and agency matter more than certifications• How to rethink workflows instead of chasing shiny AI tools• The FLIP framework: Focus, Leverage, Influence, and PowerThis episode isn’t about fear. It’s about reinvention. If you’re wondering how to stay relevant, valuable, and resilient as AI reshapes work, this is the place to start.GuestNikki BaruaCo-founder, Footwork(Reinventing organizations with agentic AI)👉 Subscribe for more conversations on AI, work, and the future of technology:https://techfirst.substack.comChapters:00:00 — Work in the AI Age: what happens to your job?01:05 — Layoffs, AI, and why this cycle feels different02:55 — “Don’t let AI have the last laugh”04:45 — Profitable companies cutting jobs: what’s really happening06:40 — The next 18–24 months: compression before reinvention08:30 — AI’s impact on young workers and early careers10:00 — What should you be doing right now?11:20 — Why surface-level AI use won’t save your job12:40 — The rise of the “agentic human”14:20 — From doing to directing: humans + machines as partners15:55 — Why certifications and training aren’t enough17:10 — High-agency people win in the AI age18:35 — The FLIP framework: Focus and identity20:00 — Leverage: compounding capacity beyond automation21:20 — Influence: trust, authenticity, and scaled impact22:25 — Power: upgrading your personal operating system23:40 — Two shifts that make this AI revolution different25:05 — Tools vs workflows: where most people get it wrong26:25 — The real blocker: old identities and fear of change27:40 — Three steps to stay relevant in the AI age28:40 — Final thoughts + wrap-up

What if someone actually built TARS from Interstellar—and discovered it really could work?In this episode of TechFirst, host John Koetsier sits down with Aditya Sripada, a robotics engineer at Nimble, who turned a late-night hobby into a serious research project: a real, working mini-version of TARS, the iconic robot from Interstellar.Aditya walks through why TARS’s strange, flat form factor isn’t just cinematic flair—and how it enables both walking and rolling, one of the most energy-efficient ways for robots to move. We dive into leg-length modulation, passive dynamics, rimless wheel theory, and why science fiction quietly shapes real robotics more than most engineers admit.Along the way, Aditya explains what he learned by challenging his own assumptions, how the project connects to modern humanoid and warehouse robots, and why reliability—not flash—is the hardest problem in robotics today. He also previews his next ambitious project: building a real-world version of Baymax, exploring soft robotics and safer human-robot interaction.This is a deep, accessible conversation at the intersection of science fiction, physics, and real-world robotics—and a reminder that sometimes the ideas we dismiss as “impossible” just haven’t been built yet.⸻GuestAditya SripadaRobotics Engineer, NimbleResearcher in legged locomotion, humanoids, and unconventional robot form factors⸻If you enjoyed this episode, subscribe for more deep dives into technology, robotics, and innovation:👉 https://techfirst.substack.com⸻Chapters:00:00 – TARS in Real Life: Why Interstellar’s Robot Still Fascinates Us01:00 – Why Building TARS Seemed Physically Impossible02:00 – From Weekend Hobby to Serious Robotics Research03:00 – How Science Fiction Quietly Shapes Real Robot Design04:00 – Walking vs Rolling: Why TARS Uses Both05:00 – Why Simple Robots Can Beat Complex Humanoids06:00 – Turning Legs into a Wheel: The Rolling Mechanism Explained07:00 – Leg-Length Modulation and Passive Dynamics08:00 – Inside the Actuators: Degrees of Freedom and Compact Design09:00 – Why TARS’s Arms Don’t Really Make Sense10:30 – Lessons Learned: Never Dismiss “Impossible” Ideas12:00 – Rimless Wheels, Gaits, and Robotics Theory13:00 – What This Project Taught Him at Nimble14:00 – What “Super-Humanoid” Robots Actually Mean15:30 – Why Reliability Matters More Than Flashy Demos16:30 – TARS as a Research Platform, Not a Product17:30 – From TARS to Baymax: Exploring Soft Robotics19:00 – Can We Build Safer, Friendlier Humanoid Robots?20:30 – What’s Next: Recreating Baymax in Real Life21:30 – Final Thoughts and Wrap-Up

AI is already reshaping the workforce. What about teenagers?Turns out, they might be more impacted than anyone else. After all, they're usually in low-skill entry-level jobs that AI can replace. The problem ... teens are losing their first experience with working, making money, and establishing an identity outside of their homes.In this episode of TechFirst, host John Koetsier speaks with Karissa Tang, a high school senior and UCLA research assistant, about her new study on how AI will impact teen employment. While most workforce studies focus on adults, Karissa analyzed the top 10 most popular teen jobs from cashiers to fast food workers and found something alarming: AI could reduce teen employment by nearly 30% by 2030.We dig into:• Which teen jobs are most vulnerable to AI and automation• Why cashiers and fast-food counter workers are hardest hit• The role of self-checkout, kiosks, and robots like Flippy• Which teen jobs appear safest (for now)• Why teens may be even more exposed to AI than adults• What schools, policymakers, and teens themselves can do nextThis is a must-watch conversation for parents, students, educators, and policymakers trying to understand how AI is reshaping early work experiences—and what it means for the next generation.🎙 GuestKarissa Tang• Founder, Booted (board games company)• Research Assistant, UCLA• Former Intern, NSV Wolf Capital• High school senior and author of a 20-page research paper on AI & teen employment📌 Subscribe & Stay AheadIf you want clear, thoughtful analysis on AI, technology, and the future of work, subscribe to TechFirst:👉 https://techfirst.substack.com00:00 – Will AI Kill Teen Jobs?01:35 – Why a Teen Studied Teen Employment03:10 – The Shocking 30% Job Loss Prediction05:10 – Top 10 Teen Jobs Most at Risk07:20 – Cashiers, Kiosks, and Self-Checkout09:40 – Fast Food, Retail, and AI Displacement12:15 – Which Teen Jobs Are Safest from AI15:05 – Robots Like Flippy and the Future of Cooking Jobs18:00 – Why Teen Jobs Are More Vulnerable Than Adult Jobs21:40 – The Importance of Human Interaction at Work25:10 – What Inspired the Research Study29:30 – How the Data and Methodology Worked33:40 – What Teens Can Do to Stay Employable37:30 – Skills, AI Literacy, and Creating New Opportunities41:00 – Final Thoughts on the Future of Teen Work

Are we about to create real life Terminators? Humanoid robots built for war?In this episode of TechFirst I talk with Sankaet Pathak, founder and CEO of Foundation, a California-based humanoid robot company that is not afraid of the defense market. We dig into why he is building humanoid robots that can work three shifts a day, how they plan to scale from dozens of robots to tens of thousands, and why he believes humanoid robots will one day build bases in Antarctica and cities on the moon.We also dive deep into military use cases. From logistics and infrastructure to “first body in” building breach operations, we explore how humanoid robots could change asymmetric warfare, deterrence, and who wins future conflicts.In this episode• Why humanoid robots are the next strategic advantage for countries and companies• How Foundation went from zero to a working production robot in about 18 months• The hardware secrets behind Phantom: actuators, efficiency, and safety• Why their robots can run almost 24 hours a day, three shifts at a time• The master plan: Antarctic bases, moon cities, and infinite robot labor• Why Sankaet thinks home robots should feel like a “genie in a bottle”• How humanoid robots may enter military operations and what that means for war• Whether robot soldiers lead to dominance, stalemate, or new forms of peaceGuest: Sankaet Pathak, founder and CEO of FoundationWebsite: https://foundation.botSubscribe to my Substack:https://techfirst.substack.com00:00 – Are we about to build real life Terminators?00:55 – Meet Sankaet Pathak and Foundation02:08 – How Foundation built a production humanoid in 18 months04:17 – Scaling plan: 40 robots today, 10,000 next year, 40,000 after06:11 – Why manufacturing is still mostly manual and what they learned from Tesla09:31 – The Foundation master plan: Antarctica, the moon, and infinite labor14:21 – Phantom specs: size, strength, payload, and real factory work15:36 – Actuators as robot muscles and why backdrivability matters18:41 – Running three shifts a day and solving heat and durability21:01 – Robot hands today and the tendon driven hands of tomorrow23:40 – Why home robots should feel like a “genie in a bottle”25:51 – Why the military needs humanoid robots27:54 – Dangerous, boring, and impossible jobs robots should take over29:22 – Drones, costs, and asymmetric warfare32:18 – First body in and robots that can pull the trigger33:16 – The future of war as “video game” and who wins34:49 – Peace through strength and 100,000 robots as deterrent35:22 – Final thoughts and what comes next for Foundation

Is AI the secret sauce that lets the West deglobalize supply chains and bring factories back home?In this episode of TechFirst, I talk with Federico Martelli, CEO and cofounder of Forgis, a Swiss startup building an industrial intelligence layer for factories. Forgis runs “digital engineers” — AI agents on the edge — that sit on top of legacy machinery, cut downtime by about 30%, and boost production by roughly 20%, without ripping and replacing old hardware.We dive into how AI agents can turn brainless factory lines into adaptive, self-optimizing systems, and what that means for reshoring production to Europe and North America.In this episode, we cover:• Why intelligence is the next geopolitical frontier• How AI agents can reshore manufacturing without making it more expensive• Turning old, offline machines into data-driven, optimized systems• The two-layer model: integration first, vertical intelligence second• Why most manufacturing AI projects fail at integration, not algorithms• How Forgis raised $4.5M in 36 hours and chose its lead investor• Lean manufacturing 2.0: adding real-time data and AI to Toyota-style processes• Why operators stay in the loop (and why full autonomy is a bad idea… for now)• Rebuilding industrial ecosystems in Europe and North America, industry by industry• What Forgis builds next with its pre-seed round and where industrial AI is headedGuest:👉 Federico Martelli, CEO & cofounder, Forgis (industrial intelligence for factories)🔗 More on Forgis: https://forgis.com/Host:🎙 John Koetsier, TechFirst podcast🔎 techfirst.substack.comIf you enjoy this conversation, hit subscribe, drop a comment about where you think factories of the future will live, and share this with someone thinking about reshoring or industrial AI.00:00 – Intro: AI, deglobalization, and the battle for industrial power01:20 – Why intelligence is the next geopolitical frontier02:13 – Applying AI agents to legacy machinery (not just new robots)03:10 – Integration first, intelligence second: the “digital engineers” layer03:58 – Early results: +20% production, –30% downtime05:39 – The Palantir-style model: deep factory work, then recurring licenses06:28 – Raising $4.5M in 36 hours and choosing Redalpine08:17 – Lean manufacturing, Toyota, and giving operators superpowers (not replacing them)10:18 – Big picture: reshoring production to Europe, the US, and Canada12:48 – Competing with China’s dense manufacturing ecosystems15:29 – What Forgis’ digital engineers actually do on the shop floor17:06 – How Forgis will use the pre-seed round: sales, product, then tech18:32 – Flipping the traditional stack: sales → product → tech19:22 – Wrap-up and what’s next for industrial intelligence

AI agents can already write code, build websites, and manage workflows ... but they still can’t pay for anything on their own. That bottleneck is about to disappear.In this episode of TechFirst with John Koetsier, we sit down with Jim Nguyen, former PayPal exec and cofounder/CEO of InFlow, a new AI-native payments platform launching from stealth. InFlow wants to give AI agents the ability to onboard, pay, and get paid inside the flow of work, without redirects, forms, or a human typing in credit card numbers.We talk about: • Why payments — not intelligence — are the missing link for AI agents • How agents become a new kind of customer • What guardrails and policies keep agents from spending all your money • Why enterprises will need HR for agents, budgets for agents, and compliance systems for agents • The future of agent marketplaces, headless ecommerce, and machine-speed commerce • How InFlow plans to become the PayPal of agentic systemsIf AI agents eventually hire, fire, transact, and manage entire workflows, someone has to give them wallets. This episode explores who does it, how it works, and what it means for the economy.👀 Full episode transcript + articles at: https://johnkoetsier.com🔎 Deeper insight in my Substack at techfirst.substack.com🎧 Subscribe to the podcast on any audio platforms00:00 — AI agents can’t pay yet01:00 — Why agents need financial capabilities02:45 — Developers as the first use case04:15 — Agents that build AND provision software06:00 — Agents as real customers with budgets07:30 — Payments infrastructure is the missing layer09:00 — Machine-speed commerce and GPU allocation10:15 — From RubyCoins to PayPal to agentic payments12:00 — Policy guardrails: the child debit card analogy14:00 — Accountability: every agent must be “sponsored”15:00 — HR, finance, and compliance systems for agents16:45 — Agent marketplaces and future gig platforms18:15 — Headless commerce: ghost kitchens for AI agents20:00 — Agents are the new apps21:15 — Amazon pushback and optimizing for revenue22:45 — Why agent-optimized platforms will emerge23:30 — Voice commerce, invisible ordering, and wallets24:15 — Final thoughts: building the rails for agent commerce

Are we ready for a world where everything is smart? Not just phones and apps, but buildings, robots, and delivery bots rolling down our streets?Windows ... doors ... maybe even towels. And don't forget your shoes.In this episode of TechFirst, I talk with Mat Gilbert, director of AI and data at Synapse, about physical AI: putting intelligence into machines, devices, and environments so they can sense, reason, act, and learn in the real world.We cover why physical AI is suddenly economically viable, how factories and logistics centers are already using millions of robots, the commercial race to build useful humanoids, why your home is the last frontier, and how to keep physical AI safe when mistakes have real-world consequences.In this episode:• Why hardware costs (lidar, batteries) are making “AI with a body” possible• How Amazon, FedEx, Ford, and others are already deploying physical AI at scale• The humanoid robot race: Boston Dynamics, Figure AI, Tesla, and more• Why home robots are so hard, and the “coffee test” for general humanoid intelligence• Physical AI in agtech, healthcare, and elder care• Safety, simulation, and why physical AI can’t rely only on probabilistic LLMs• Human–robot teaming and how to build trust in messy, real-world environments• What we can expect by 2026 and beyond in service robots and smart spaces00:00 – Giving AI a body: why physical AI is becoming viable01:00 – Where we are today: factories, logistics, and Amazon’s million robots03:30 – The software layer: coordinating robots, routing, and warehouse intelligence06:00 – Cloud vs edge AI: latency, cost, and why intelligence is moving to the edge10:00 – Humanoid robots: bets from Boston Dynamics, Figure AI, and Tesla14:00 – Home robots as the last frontier and the “coffee test” for generality17:00 – Beyond factories: agtech, carbon-killing farm bots, and healthcare use cases18:30 – Elder care, hospital robots, and amplifying human caregivers20:00 – Foundation models for robotics, simulation, and digital twins21:00 – Why physical AI safety is different from digital AI safety22:30 – Layers of safety, shutdown zones, and cyber-physical security risks24:30 – Human–robot teaming, trust, and communicating intent26:00 – What’s coming by 2026: service robots, delivery bots, and smart spaces28:00 – Delivery robots, drones, and physical AI in everyday environments29:00 – Closing thoughts on living in a world full of physical AI