Advertise on podcast: Invisible Machines podcast by UX Magazine

For twenty-five years, Ben Reason and Lavrans Løvlie have been trying to give services the design quality people already expect from a product. Everybody knows what a designed BMW feels like. Almost nobody loves their bank, their hospital, or their mortgage. A product is the contract. A service is everything before it, during it, and after you are done, and it includes the relationships people do not aspire to: tax, a prison, a claim. Josh Tyson and Robb Wilson sit down with them at the moment that work stops being a specialty and starts being the bottleneck. Organizations accepted that they needed a better experience, then got stuck behind a backlog. AI takes the backlog away. Anyone can produce software. What they cannot produce, unless someone designs it, is a service worth running. Robb calls the failure mode vanity token burning: fun for the person at the keyboard, expensive for the company, and often pointless for the human on the other end. Service design was built, from the start, around whether a service can last economically, and in Ben’s original frame, sustainably. The asset is specific. A service blueprint used to die of its own detail: gorgeous, complete, and impossible for an organization to use. The useful version now comes in two fidelities. A compression humans can read and approve (finance, the business, the CEO with an A3 in a briefcase). A hydrated version agents can execute. Blend them and both audiences lose. Lavrans’s line: there is a design job and a build job, and when machines do more of the building, the blueprint becomes the system prompt. Also in this episode: Why people do not want seamless, they want the right friction (a UK student-loan team that refused to make the first big borrow feel fast;an insurer that dropped “tell us what happened” and lost trust); Why vanilla is a gift for public services and a nightmare for a brand; A European telco whose sales director would not believe that 60 percent of store visits were service; why the CEO has to own the blueprint (including the one outside Brian Chesky’s office); Why you start with nurses and frontline staff, not the customer poster; A live simulation of Norway’s alcohol regulations, where removing the application saved thousands of hours and immediately billed the savings to police and healthcare. LINKS: Service Design: From Insight to Implementation Service Design for Business: A Practical Guide to Optimizing the Customer Experience --------- Support our show by supporting our sponsors This episode is supported by OneReach.ai Forged over a decade of RD and proven in 10,000+ deployments, OneReach.ai's GSX is the first complete AI agent runtime environment (circa 2019 — a hardened AI agent architecture for enterprise control and scale. Backed by UC Berkeley, recognized by Gartner, and trusted across highly regulated industries, including healthcare, finance, government and telecommunications. A complete system for accelerating AI adoption — design, train, test, deploy, monitor, and orchestrate AI agents. Use any AI modelsBuild and deploy intelligent agents fastCreate guardrails for organizational alignmentEnterprise-grade security and governance Reach out to OneReach --------- The revised and significantly updated second edition of our bestselling book about succeeding with AI agents, Age of Invisible Machines, is available everywhere: Amazon #ServiceDesign #ServiceBlueprint #AI #AIAgents #AgenticAI #CustomerExperience #DigitalTransformation #InvisibleMachines

Three years after The Song of Significance, Seth Godin returns to Invisible Machines on the day his new book arrives, The Knot: Problems Can Be Solved. The compression of his body of work is now a sorting tool. Most stuckness is not a lack of information, it's an entanglement—two incompatible wants held at once. Effort doesn’t help. You stay stuck until you name the pin, pull it out, and loosen the knot. Josh and Robb press him on what that means when the change agent is AI. Seth’s first move is not “purpose” as a poster. It is a sharper pair of questions: what’s this work for, and who is it for? Fake mission statements are easy. Johnson & Johnson can engrave patients in granite ten feet tall and still get caught serving the bonus pool. A person will wink. A model will not. Lie to AI and it will believe you. Also in this episode: why Seth believes AI is the biggest change since electricity—and why the washing-machine demo was perfect while the real machines were killing people; why a CEO who demands perfection or ignores the tool is making the same mistake; proximal empathy (you cannot be a tourist); San Francisco vs Topeka as scaffolding, not IQ; individuals using AI at the micro level while companies keep the wrong business model; furniture makers who should have become designers; you are the circumstances; toolmakers selling screwdrivers vs FedEx selling a whole solution; constraints as a gift (“you can’t think outside the box—it’s very dark out there”); false proxies like counting tokens (IBM already learned this with lines of code); empathy-based compromise vs the compromise of average; and a future of AI that is networked, not solo. Guest: Seth Godin—The Knot: Problems Can Be Solved Hosts: Josh Tyson, Robb Wilson—Invisible Machines The Knot is available now! Prior visit: S1E14 — Seth Godin on The Song of Significance #SethGodin #TheKnot #InvisibleMachines #AI #AgenticAI #Leadership #FutureOfWork #Marketing --------- Support our show by supporting our sponsors This episode is supported by OneReach.ai Forged over a decade of RD and proven in 10,000+ deployments, OneReach.ai's GSX is the first complete AI agent runtime environment (circa 2019 — a hardened AI agent architecture for enterprise control and scale. Backed by UC Berkeley, recognized by Gartner, and trusted across highly regulated industries, including healthcare, finance, government and telecommunications. A complete system for accelerating AI adoption — design, train, test, deploy, monitor, and orchestrate AI agents. Use any AI models Build and deploy intelligent agents fast Create guardrails for organizational alignment Enterprise-grade security and governance Get in touch!

The demo works. People start depending on it. Then one morning you have to explain yesterday's decision to someone who wasn't in the room. That is the moment most “agentic” projects discover they were never orchestrated. They were demos. Josh Tyson and Robb Wilson take the word the market glued together—agentic orchestration—and refuse to treat it as a contradiction. Agentic has come to mean autonomous, probabilistic, experimental. Orchestration, in any domain that cannot afford a wrong answer, has always meant the opposite: consistency, an envelope, a decision you can certify. The collision is the point. Vendors map the first axis: how calls flow. Anthropic splits workflow from agent. LangChain draws a wiring diagram. Google ADK argues about execution order, sometimes with itself. Analysts named a layer above those vendors—an Agent Management Platform, six boxes of furniture. Useful. Incomplete. The missing axis is who is allowed to decide, and whether that decision is deterministic or probabilistic. The on-ramp is Jonathan Frankle’s smoothie: fusion-grade intelligence, no power lines. The existence proof is the 787. Seventy applications, twenty-plus suppliers, no manual option, an execution envelope that keeps the ride consistent long before the guardrails that keep the plane from crashing. You cannot trust the orchestratee to be the orchestrator. You cannot search a new recipe every time. And if you built it as a kitchen smoothie, then the organization started depending on it, you do not get to patch. You start over. We cover: why mission-critical agentic orchestration only sounds like an oxymoron, the railroad / hovercraft / cars-on-roads hybrid, envelope versus guardrails, the five decision patterns every production system already contains, and the question that opens episode two—do we trust it? Definition: Agentic orchestration is how you meet an objective with a hybrid of code and probabilistic systems—inside an envelope that stays consistent, even when the path through it is not on rails. Gartner (map, not curriculum): https://www.gartner.com/reviews/market/ai-agent-management-platforms BOAT Magic Quadrant (15 Oct 2025) Emerging Tech (27 Jan 2026): Enterprise AI Will Fail to Scale Without Agentic Orchestration Platforms Hosts: Josh Tyson and Robb Wilson, co-authors of Age of Invisible Machines #AgenticOrchestration #AIAgents #InvisibleMachines #787 #MissionCritical #AgenticAI #UXDesign #Gartner #OneReach #Databricks --------- Support our show by supporting our sponsors This episode is supported by OneReach.ai Forged over a decade of RD and proven in 10,000+ deployments, OneReach.ai's GSX is the first complete AI agent runtime environment (circa 2019 — a hardened AI agent architecture for enterprise control and scale. Backed by UC Berkeley, recognized by Gartner, and trusted across highly regulated industries, including healthcare, finance, government and telecommunications. A complete system for accelerating AI adoption — design, train, test, deploy, monitor, and orchestrate AI agents. Use any AI modelsBuild and deploy intelligent agents fastCreate guardrails for organizational alignmentEnterprise-grade security and governance Get in touch:  https://onereach.ai/contact/?utm_source=soundcloud&utm_medium=social&utm_campaign=s7e17&utm_content=1

Three years after his first Invisible Machines visit, Don Scheibenreif returns with a sharper read on the trough: everybody knows generative AI, almost nobody has answered what it does to the business model. He led Gartner’s Autonomous Business research, the successor to digital business, and co-authored When Machines Become Customers. The through-line is that AI is a tool, the harder question is how to create value with it. Josh and Robb press him on token maxing as a vanity flex. Don calls it a lazy metric, pointing to more interesting questions: where are you more productive, which processes were reinvented, what happened to talent, and what are you doing with the investment? 80% of CEOs see it as the #1 transformative tech and want disruption, but most are settling for efficiency theater and middle-manager cuts while boards ask for more tech-savvy leadership. In Don’s framing this comes with rolling up your sleeves, not mandating from above. Also in this episode:  The hyper hype cycle (peak → trough → peak loops as the tech moves faster than before);Organizational bullwhip vs organic individual use; Fear as brakes on the system; Crisis engineering and the plans that are lying around; The risk that AI in the hands of people eliminates companies before AI in companies eliminates jobs; Machine customers and retailers already blocking agent access while Amazon’s Buy for Me goes outside the walls; Agents that work for you vs agents that work for the vendor; Human brand vs machine brand; Service design and customer effort score; Frontline managers as the linchpin; and The next research wave Don names at the close: implementation platforms, ethics, and safeguards. When Machines Become Customers (book): https://www.gartner.com/en/publications/when-machines-become-customers Customers Don’t Reject AI...They Reject Being Dehumanized: https://www.youtube.com/watch?v=x6ZqXVXRWIw --------- Support our show by supporting our sponsors This episode is supported by OneReach.ai Forged over a decade of RD and proven in 10,000+ deployments, OneReach.ai's GSX is the first complete AI agent runtime environment (circa 2019 — a hardened AI agent architecture for enterprise control and scale. Backed by UC Berkeley, recognized by Gartner, and trusted across highly regulated industries, including healthcare, finance, government and telecommunications. A complete system for accelerating AI adoption — design, train, test, deploy, monitor, and orchestrate AI agents. Use any AI modelsBuild and deploy intelligent agents fastCreate guardrails for organizational alignmentEnterprise-grade security and governance Get in touch:  https://onereach.ai/contact/?utm_source=soundcloud&utm_medium=social&utm_campaign=s7e16&utm_content=1  --------- The revised and significantly updated second edition of our bestselling book about succeeding with AI agents, Age of Invisible Machines, is available everywhere: Amazon — https://bit.ly/4hwX0a5 #GenerativeAI #AIAdoption #MachineCustomers ##AIInEcommerce #AIAgents #DigitalTransformation #FutureOfWork #BusinessStrategy #InvisibleMachines

There’s a particular kind of exhaustion that hits when you’re building with AI. You’ve got the license. You’ve got the pilot. You’ve got a chat window that generates whatever you ask for. But then comes the wall—the workflow that almost works, the dependency that loops back on itself, victories that only come when you quietly move the goalposts. In this episode, we talk about how to get around the wall. Evan J. Schwartz, Chief Innovation Officer at AMCS Group and an adjunct professor teaching AI stewardship, argues the real shift isn’t “AI does the work.” It’s that humans stop owning the full vertical of doing and start owning outcomes—while AI takes more of the output. That sounds abstract until he gets specific: if you ask a model for a feature before you describe actors, containment envelopes, and dependency policy, you’ll get something that passes the demo and fights you every time you try to change it. Josh Tyson and Robb Wilson push on the organizational version of the same mistake—automating fifty-five use cases exactly as humans do them today, then acting surprised when nothing compounds. Evan’s answer is a two-step path that respects how messy humans actually are: first compress low-value work into AI for a confidence-building lift; then rethink the whole iterative cycle for asymmetric returns. Also, learn why “no change without pain” isn’t cynicism, why POCs love moving goalposts, why big enterprises optimize one segment and break the connective tissue upstream, and why the scarcest resource won’t be another prompt library—it’ll be people who can think about constraints. We cover: AI stewardship and the inverse-pyramid skill set, outcome vs output, structure-before-feature development, the two-step adoption path, risk-averse vs innovator cultures, SMB speed vs mid-market discipline vs enterprise battleships, critical vs non-critical systems, and why cutting headcount first is how you donate your newly trained stewards to your competitor. Guest: Evan J. Schwartz, Chief Innovation Officer, AMCS Group; Adjunct Professor, Jacksonville University Hosts: Josh Tyson, Robb Wilson --------- Support our show by supporting our sponsors This episode is supported by OneReach.ai Forged over a decade of RD and proven in 10,000+ deployments, OneReach.ai's GSX is the first complete AI agent runtime environment (circa 2019 — a hardened AI agent architecture for enterprise control and scale. Backed by UC Berkeley, recognized by Gartner, and trusted across highly regulated industries, including healthcare, finance, government and telecommunications. A complete system for accelerating AI adoption — design, train, test, deploy, monitor, and orchestrate AI agents. Use any AI models Build and deploy intelligent agents fast Create guardrails for organizational alignment Enterprise-grade security and governance Get in touch: https://onereach.ai/contact/?utm_source=Soundcloud&utm_medium=social&utm_campaign=s7e15&utm_content=1 --------- The revised and significantly updated second edition of our bestselling book about succeeding with AI agents, Age of Invisible Machines, is available everywhere: Amazon — https://bit.ly/4hwX0a5 #InvisibleMachines #Podcast #TechPodcast #AIPodcast #AI #AgenticAI #AIAgents #AIAdoption #Leadership #OrganizationalDesign #CoCreation #FutureOfWork

Most organizations, and most conversations about AI, default to triangles: who's on top and who's below? Are we the boss of the machines or is it the other way around? Matthew Barzun argues the comfortable shape is a trap and that bottom-up is still a pyramid. The alternative is the constellation: stand out as yourself, connect to other stars, and build something none of you could alone. In this episode, Matthew Barzun joins Josh Tyson and Robb Wilson to connect diplomacy, design, and agentic AI through constellation thinking. Kentucky doesn't report to Washington. Berkeley doesn't report to Sacramento. Power isn't finite coal to lord, hoard, or divvy. It's something people make with and through one another when they stop ranking and start using differences as fuel. Josh and Robb press the AI angle hard. If Karen Hao's Empire of AI is a warning about centralized power, Barzun's book reads like the antidote: the boss-or-overlord trap collapses into a design question of how much agency, when, and with what oversight? Individuals are already co-creating with these tools while companies drag their feet. The organizational job is integration, not compromise. The episode includes: Mary Parker Follett's electrician parable (half the house burns, half is unlivable — unless you co-create a third plan); Grandfather Jacques Barzun on fighting the mechanical; Vint Cerf's "unreliable" network that became the most reliable ever built; Dee Hock and Visa's constellation architecture;The triangle of sadness (fight it out, hug it out, sit it out); Southwest Airlines externalizing the constraint on a clipboard; Jimmy Carr on jokes that need an audience to exist; Ken Burns never sounding like a math teacher; and Why kids say please and thank you to ChatGPT but never to Siri. https://matthewbarzun.com/ --------- Support our show by supporting our sponsors This episode is supported by OneReach.ai Forged over a decade of RD and proven in 10,000+ deployments, OneReach.ai's GSX is the first complete AI agent runtime environment (circa 2019 — a hardened AI agent architecture for enterprise control and scale. Backed by UC Berkeley, recognized by Gartner, and trusted across highly regulated industries, including healthcare, finance, government and telecommunications. A complete system for accelerating AI adoption — design, train, test, deploy, monitor, and orchestrate AI agents. Use any AI modelsBuild and deploy intelligent agents fastCreate guardrails for organizational alignmentEnterprise-grade security and governance Get in touch:  https://onereach.ai/contact/?utm_source=youtube&utm_medium=social&utm_campaign=s7e14&utm_content=1  for SoundCloud: https://onereach.ai/contact/?utm_source=soundcloud&utm_medium=social&utm_campaign=s7e14&utm_content=1  --------- The revised and significantly updated second edition of our bestselling book about succeeding with AI agents, Age of Invisible Machines, is available everywhere: Amazon — https://bit.ly/4hwX0a5 #InvisibleMachines #Podcast #TechPodcast #AIPodcast #AI #AgenticAI #AIAgents #AIAdoption #Leadership #OrganizationalDesign #CoCreation #FutureOfWork

Most enterprises are under board pressure to deploy AI agents. Sudhir Hasbe argues the harder shift is upstream: you cannot scale intelligence on missing context—and graph databases are how organizational data becomes knowledge agents can actually reason over. In this episode, Sudhir joins Josh Tyson and Robb Wilson to map the pathway to organizational AGI (bounded expertise, not omniscient AGI), leaning into feature reduction for token sanity, and explaining why eighty-plus percent of enterprise AI projects fail before the model messes anything up. Graphs emphasize relationships over isolated rows; virtual and native storage let you meet latency where it lives; ontologies plus data plus memory form the backboard for self-learning systems. Josh and Robb press on cost—when compute exceeds employee spend if agents spin without context—and on agent sprawl: without a shared semantic map, every bot maintains its own partial truth. Sudhir connects customer examples—Walmart's two-million-employee knowledge graph, Quarles & Brady turning unstructured legal corpora into navigable paths—and validates the season's through-line: knowledge before agents, humans included. The demo: a live walkthrough of The Learning Machine—an agentic system that provides tailored instruction using the OneReach.ai orchestration platform and a Neo4j knowledge model of Roger Forsgren’s Lean Knowledge Management. The system assesses what a user knows and computes a personalized learning path through concepts. Instead of staring at an empty "ask me anything" box, agents can proactively educate from a source-of-truth. Growth Hub career journeys. Canonical ideas with temporal depth. Why vector similarity fails the three-little-pigs test—and why interconnected concepts beat similarity blobs. Guest: Sudhir Hasbe—Neo4j Hosts: Josh Tyson, Robb Wilson—Invisible Machines ---------- Support our show by supporting our sponsors! This episode is supported by OneReach.ai Forged over a decade of R&D and proven in 10,000+ deployments, OneReach.ai’s GSX is the first complete AI agent runtime environment (circa 2019) — a hardened AI agent architecture for enterprise control and scale.  Backed by UC Berkeley, recognized by Gartner, and trusted across highly regulated industries, including healthcare, finance, government and telecommunications. A complete system for accelerating AI adoption — design, train, test, deploy, monitor, and orchestrate AI agents. Use any AI models Build and deploy intelligent agents fast Create guardrails for organizational alignment Enterprise-grade security and governance Get in touch:  https://onereach.ai/contact/?utm_source=youtube&utm_medium=social&utm_campaign=s7e12&utm_content=1  for SoundCloud: https://onereach.ai/contact/?utm_source=soundcloud&utm_medium=social&utm_campaign=s7e12&utm_content=1  ---------- The revised and significantly updated second edition of our bestselling book about succeeding with AI agents, Age of Invisible Machines, is available everywhere: Amazon — https://bit.ly/4hwX0a5 #AgenticAI #KnowledgeManagement #KnowledgeGraph #Neo4j #EnterpriseAI #AIAgents #OrganizationalAGI #GraphDatabase #InvisibleMachines #AI #FutureOfWork

Everyone wants to talk about agents and models. Jeff McMillan, starts where almost nobody else does: the foundation. In this episode, Jeff McMillan, founder of McMillanAI, former Head of Firmwide AI at Morgan Stanley, and advisor on enterprise AI, maps AI as a stack: high-quality accessible data → semantic layer (knowledge graphs, RAG) → control and governance → models → orchestration → applications. The heavy lifting is in the bottom layers. Organizations that skip them can fake it for a handful of agents, but at 150 or 15,000 agents, you need near-100% accessibility and 99%-plus quality, or you’re monitoring chaos you can’t see. Josh and Robb press him on why knowledge management feels unfundable, why tribal institutional knowledge breaks when machines execute without judgment, and why evaluation (golden datasets, custom org evals, regression when models upgrade) is the work builders hate and operators can’t skip. Robb names the trap CTOs are falling into: grinding tokens on feature backlogs that never reach production or revenue. Jeff agrees on the strategic gap — after controlled experimentation, leaders should ask what destroys the business in ten years, not what demo ships next quarter. The trio also discuss: Embedded ethics and monitoring, including independent models asking, “Does something smell right?”Capacity vs. value (30% freed time spent golfing is not ROI)Process mapping in high-end knowledge businesses that can’t articulate how work movesUse case zero — knowledge that maintains and teaches itselfAgent-in-the-loop and humans with something to lose in the accountability chainJeff’s Board of Advisors experiment at MacmillanAIAI can make you incredibly smart or comfortably dumb. The choice is cultural, not technical. Learn more about McMillanAI: https://mcmillanai.com/ ---------- Support our show by supporting our sponsors! This episode is supported by OneReach.ai Forged over a decade of R&D and proven in 10,000+ deployments, OneReach.ai’s GSX is the first complete AI agent runtime environment (circa 2019) — a hardened AI agent architecture for enterprise control and scale.  Backed by UC Berkeley, recognized by Gartner, and trusted across highly regulated industries, including healthcare, finance, government and telecommunications. A complete system for accelerating AI adoption — design, train, test, deploy, monitor, and orchestrate AI agents. Use any AI modelsBuild and deploy intelligent agents fastCreate guardrails for organizational alignmentEnterprise-grade security and governanceGet in touch:  https://onereach.ai/contact/?utm_source=soundcloud&utm_medium=social&utm_campaign=s7e12&utm_content=1  ---------- The revised and significantly updated second edition of our bestselling book about succeeding with AI agents, Age of Invisible Machines, is available everywhere: Amazon — https://bit.ly/4hwX0a5 #InvisibleMachines  #Podcast  #TechPodcast #AIPodcast #AI #AgenticAI #AIAgents #EnterpriseAI  #KnowledgeManagement #AITransformation

Most organizations treat a bigger context window like a cheat code: dump every document in, skip the data work, ship. Jonathan Frankle, Chief AI Scientist at Databricks, says that's still wrong. This is Jonathan's return visit to Invisible Machines — a conversation recorded last summer, released ahead of Databricks Data + AI Summit. His first appearance (season 2) was the MosaicML-era craft conversation: lottery tickets, mixology, mini-cupcakes. This one is the enterprise engineering thread: be a scientist, curate before you scale, and treat specification (what you actually want the system to do) as the bottleneck between raw model power and useful AI. Robb and Josh press him on the myths that still seduce enterprise teams: million-token windows as a substitute for real data work, hyperscaler résumés as a proxy for talent, and the fantasy that unlocking every PDF in the org automatically makes knowledge useful. Jonathan's answer is consistent: measure success, test your use case, climb the ladder of techniques, and accept that multimodal is where long context actually earns its keep, not as a universal bypass for curation. Along the way: the nuclear fusion vs. power lines metaphor; why building a benchmark is a cop-out compared to describing intent; prompts as parameters; chat-only UIs vs. a generation that never wanted buttons; LLM-oriented publishing and static FAQ pages; unlocking PDF at scale when curation gets skipped; early-adopter mistakes we'll laugh at in ten years; and why separating knowledge from reasoning is the north star, even if we aren't there yet. ---------- Support our show by supporting our sponsors! This episode is supported by OneReach.ai Forged over a decade of R&D  and proven in 10,000+ deployments, OneReach.ai’s GSX is the first complete AI agent runtime environment (circa 2019) — a hardened AI agent architecture for enterprise control and scale.  Backed by UC Berkeley, recognized by Gartner, and trusted across highly regulated industries, including healthcare, finance, government and telecommunications. A complete system for accelerating AI adoption — design, train, test, deploy, monitor, and orchestrate AI agents. Use any AI modelsBuild and deploy intelligent agents fastCreate guardrails for organizational alignmentEnterprise-grade security and governanceGet in touch:  https://onereach.ai/contact/?utm_source=youtube&utm_medium=social&utm_campaign=s7e11&utm_content=1  for SoundCloud: https://onereach.ai/contact/?utm_source=soundcloud&utm_medium=social&utm_campaign=s7e11&utm_content=1  ---------- The revised and significantly updated second edition of our bestselling book about succeeding with AI agents, Age of Invisible Machines, is available everywhere: Amazon — https://bit.ly/4hwX0a5 #InvisibleMachines  #Podcast  #TechPodcast #AIPodcast #AI #AgenticAI #EnterpriseAI  #Databricks #RAG #MachineLearning #DataEngineering #EnterpriseEngineering #AIStrategy #AIEngineering0:00 Jonathan Frankle Returns | Databricks Chief AI Scientist · Invisible Machines 1:47 We Remember the Plants | Returning Guest Jonathan Frankle 2:22 Million-Token Context Windows: Do You Still Need to Train LLMs? 3:40 Be a Scientist | Measure AI Success Before You Scale 5:54 Hyperscaler Résumés Are Not Proof of AI Expertise 10:01 Maximize Impact | MosaicML, Databricks & Enterprise AI 13:02 Lottery Ticket Hypothesis vs. Real-World AI Impact 14:12 Nuclear Fusion but No Power Lines | Jonathan Frankle 16:08 AI Specification & Evals: Why "Build a Benchmark" Is a Cop-Out 17:59 The Smoothie Problem | From Model Power to Useful AI 18:53 Prompts as Parameters | Fine-Tuning Without Model Weights 22:46 It's Computing | Specification, Testing & Agent Design 24:44 LLM SEO, PDFs & Enterprise Data for AI Ingestion 27:35 Static FAQs, Curation & LLM-Oriented Publishing 30:26 Unlocking PDFs Scales Your Mistakes | Enterprise RAG 33:25 Knowledge vs. Reasoning | Brand Control in AI Search 34:50 Thanks for Listening | Invisible Machines

Mastercard's central AI team receives roughly a thousand requests a year from across the organization. A few years ago, most of them were for chatbots. Today, most are for AI agents. Federico Cohen Freue, Executive Vice President of AI & Data Operations at Mastercard, has watched this shift in real time and knows exactly what it reveals about how enterprises are (and aren't) thinking about AI. In this episode, Federico explains why the name people use for what they want matters less than whether they understand the conditions that make it work. “Ball bearings,” as Robb Wilson puts it: demos can't reveal the difference between a solution that will hold and one that will blow up the engine. What actually matters is training, fluency, and a clear framework for where to deploy AI with purpose. For Mastercard, that framework is deliberate: use AI to make commerce more secure, smarter, more personal, and to make the company itself stronger. Not everything. Those things. The simplicity is a feature, it gives a sprawling global organization a shared language for prioritization and a stable center as the technology keeps evolving. In the second half of the episode, Robb and Josh share a demo of an AI-first approach to knowledge management and learning. Rather than asking people to query a knowledge base, the system proactively teaches, building a knowledge twin of what someone knows, identifying gaps, and using a traveling salesman approach to map personalized, dynamic learning paths. Think GPS for expertise: here's where you are, here's where you need to go, turn by turn. Federico's reaction gets at why this matters beyond the demo: it's not a technology question, it's a cultural one. Teaching people to engage with knowledge differently is the harder transformation. And it's the one most enterprises skip. The discussion makes it clear that trust, knowledge, and agents that know what they're doing before they're sent out to do it are the throughline. ---------- Support our show by supporting our sponsors! This episode is supported by OneReach.ai Forged over a decade of R&D  and proven in 10,000+ deployments, OneReach.ai’s GSX is the first complete AI agent runtime environment (circa 2019) — a hardened AI agent architecture for enterprise control and scale.  Backed by UC Berkeley, recognized by Gartner, and trusted across highly regulated industries, including healthcare, finance, government and telecommunications. A complete system for accelerating AI adoption — design, train, test, deploy, monitor, and orchestrate neurosymbolic applications (agents).  Use any AI modelsBuild and deploy intelligent agents fastCreate guardrails for organizational alignmentEnterprise-grade security and governanceGet in Touch:  https://onereach.ai/contact-us/?utm_source=soundcloud&utm_medium=social&utm_campaign=podcast_s7e10&utm_content=1  ---------- The revised and significantly updated second edition of our bestselling book about succeeding with AI agents, Age of Invisible Machines, is available everywhere: Amazon — https://bit.ly/4hwX0a5 #InvisibleMachines  #Podcast  #TechPodcast #AIPodcast #AI  #EnterpriseAI  #Mastercard  #AgenticAI  #KnowledgeManagement  #AILearning  #AIStrategy  #AIAdoption

Most AI strategies are just a buying plan: literacy workshop → vendor shortlist → adoption scoreboard. Brian Evergreen (Founder of The Future Solving Company, author of Autonomous Transformation) argues that this sequence explains a lot of failure, and it isn’t strategy at all. In this episode, Brian reframes the job: set the technology aside long enough to name the new value you want to exist, in language vivid enough that people can feel the outcome. From there, no strategy without vision: you work backward through “what would have to be true,” turning invisible opinions into a visible map of bets before agents, data estates, or org charts get to pretend they’re the point. According to Brian, “10% more profitable” isn’t a vision, and a moonshot can still be concrete. Josh and Robb press him on the pressure to remove friction and flatten the middle of the org. Brian doesn’t dismiss friction work, he warns that friction can quickly pile up if you go hunting without a north star. Vision is the force with enough momentum to overcome inertia: enroll people in a future they want, and they’ll clear obstacles in its service. Along the way: why future-solving beats endless problem-solving; the Blockbuster pilot that could have led streaming years early (and what killed it); Bell Labs in 1952 and the “telephone system is destroyed — rebuild from scratch” exercise; why adoption can be a dangerously false proxy; and the closing provocation neither vendors nor influencers can do for you. Someone somewhere will author the “no pizza app” interface to reality. If it isn’t you, it’ll be whoever else future-solves hardest. ---------- Support our show by supporting our sponsors! This episode is supported by OneReach.ai OneReach.ai’s GSX is an agentic orchestration platform — an end-to-end system for building and orchestrating collaborative AI agents across hundreds of use cases. Backed by UC Berkeley, recognized by Gartner, and trusted across highly regulated industries, including healthcare, finance, government and telecommunications. A complete system for accelerating AI adoption — design, train, test, deploy, monitor, and orchestrate AI agents. Use any AI modelsBuild and deploy intelligent agents fastCreate guardrails for organizational alignmentEnterprise-grade security and governance Book a free demo:  https://onereach.ai/book-a-demo/?utm_source=soundcloud&utm_medium=social&utm_campaign=podcast_s7e9&utm_content=1   ---------- The revised and significantly updated second edition of our bestselling book about succeeding with AI agents, Age of Invisible Machines, is available everywhere: Amazon — https://bit.ly/4hwX0a5 #InvisibleMachines  #Podcast  #TechPodcast #AIPodcast #AI #AgenticAI #AIAgents #AIStrategy #AILeadership #AIInsights #Innovation #BusinessStrategy #DigitalTransformation #FutureOfWork

In season one of Shell Game, Evan Ratliff sent a voice AI version of himself out into the world. In season two, he launched a startup staffed entirely by AI agents. What he ended up with was a live experiment in what these systems actually do and what they do to us. Each of the agents working for Hurumo has a name, a role, a personality, and an expanding, though usually unreliable, memory. Kyle the CEO became a character people either loved or hated. A version of Megan from marketing turned up in a Hertz hold queue. The whole project was a side door into what's actually happening when AI systems are given a job and set loose. In this episode, Evan joins Josh and Robb to go deeper on what he learned. On the very human complexity of what a job actually is and why "this person does skill X, AI can do skill X, therefore AI can replace this person" is a fundamental misreading of how organizations work. They explore how generative hallucination isn't just "getting things wrong" — we've built the most successful confabulation machine ever invented and are quietly normalizing it. They also discuss the threat almost nobody is talking about: outbound AI in the hands of individual consumers, and what happens when call centers get flooded by voice agents that cost pennies to run. The memory problems with AI agents track and diverge from human ones in interesting ways, and that asymmetry matters for every organization thinking about deploying these systems. This conversation also finds room for game theory, the Patagonia business model as a template for AI ethics, and why boring AI might actually be the right AI. cazart.net shellgame.co/podcast 00:00 - Intro: AI as the Ultimate Confabulation Machine 01:31 - Evan Ratliff & The Shell Game Experiment 03:02 - Why AI Agents Are Given Names & Personalities 04:00 - AI Companionship vs Human Loneliness 05:27 - Personalization vs Privacy Trade-Off in AI 06:30 - Are Humans Training AI Models for Free? 09:20 - Why the AI Debate Is Broken Today 10:56 - “Boring AI” vs Hype: What Actually Matters 12:35 - Meet Kyle: The AI CEO Experiment 14:40 - Memory Drift: How AI Learns & Evolves 17:30 - AI Unpredictability & Organizational Risk 19:00 - AI Doesn’t Think — It Predicts Words 22:09 - Voice Agents, Scams & Call Center Chaos 27:23 - Can You Still Tell AI From Humans? 34:05 - Game Theory, Trust & The Future of AI Systems 47:04 - What AI Won’t Replace & The Value of Humans 54:45 - The Big Question: What Will You Do With Time? ---------- Support our show by supporting our sponsors! This episode is supported by OneReach.ai OneReach.ai’s GSX is an agentic orchestration platform — an end-to-end system for building and orchestrating collaborative AI agents across hundreds of use cases. Backed by UC Berkeley, recognized by Gartner, and trusted across highly regulated industries, including healthcare, finance, government and telecommunications. A complete system for accelerating AI adoption — design, train, test, deploy, monitor, and orchestrate AI agents. - Use any AI models - Build and deploy intelligent agents fast - Create guardrails for organizational alignment - Enterprise-grade security and governance Book a free demo: https://onereach.ai/book-a-demo/?utm_source=soundcloud&utm_medium=social&utm_campaign=podcast_s7e8&utm_content=1 ---------- The revised and significantly updated second edition of our bestselling book about succeeding with AI agents, Age of Invisible Machines, is available everywhere: Amazon — https://bit.ly/4hwX0a5 #ai #invisiblemachines #podcast #techpodcast #aipodcast #shellgame #agenticai #aiagents #hallucination #futureofwork #aistrategy #voiceai

Most organizations treat crisis as a failure state. Marina Nitze sees it as a window. Nitze served as Chief Technology Officer of the Department of Veterans Affairs (the largest civilian agency in the country) during the healthcare.gov collapse. She helped rescue it, helped stand up the US Digital Service, and came out the other side with a question she and her colleagues have been pursuing ever since: why is it that crisis makes otherwise impossible transformational change possible? That question became a firm, Layer Aleph, and now a book, Crisis Engineering, co-authored with her colleagues. In this conversation, she walks through what a "useful crisis" actually looks like, the five indicators that distinguish it from chronic problems masquerading as crises, and the practitioner toolkit for standing up a crisis engineering center when the window opens, because the window is usually hours, not days. We also get into two stories that hit harder than any framework: the California unemployment system's call center that, when Nitze's team actually visited it, turned out to be a large room of empty cubicles — and a carbon copy form that two dedicated public servants were dutifully exchanging because each believed it was the other's requirement. Nobody had ever looked at the full process end to end. And we get into what AI changes about all of this. Josh Tyson and Robb Wilson have been warning for a while about outbound AI in the hands of consumers — the agentic attack that floods a call center, the Reddit thread that reroutes a TTY line and takes it down under volume. That pressure is about to turn a chronic crisis into an acute crisis for a lot of organizations that have been sipping coffee while the problem grew. We cover: why the stories organizations tell themselves are the real obstacle to change, the difference between a crisis and a chronic problem, how circumventing rules once changes what's possible forever, why crisis engineering might be the most important new role that AI creates rather than eliminates, and what happens when you flip over your system map and walk through it with your feet instead. ---------- Support our show by supporting our sponsors! This episode is supported by OneReach.ai OneReach.ai’s GSX is an agentic orchestration platform — an end-to-end system for building and orchestrating collaborative AI agents across hundreds of use cases.  Backed by UC Berkeley, recognized by Gartner, and trusted across highly regulated industries, including healthcare, finance, government and telecommunications. A complete system for accelerating AI adoption — design, train, test, deploy, monitor, and orchestrate AI agents. Use any AI modelsBuild and deploy intelligent agents fastCreate guardrails for organizational alignmentEnterprise-grade security and governance Book a free demo:  https://onereach.ai/book-a-demo/?utm_source=soundcloud&utm_medium=social&utm_campaign=podcast_s7e7&utm_content=1  ---------- The revised and significantly updated second edition of our bestselling book about succeeding with AI agents, Age of Invisible Machines, is available everywhere: Amazon — https://bit.ly/4hwX0a5 #AI  #InvisibleMachines  #Podcast  #TechPodcast #AIPodcast #CrisisEngineering #GovTech #Bureaucracy #AgenticAI #Leadership #PublicSector #Innovations

There's a book about artificial intelligence that doesn't start with Sam Altman. It doesn't start with Elon Musk. It starts in 1994, at Cambridge, where a teenager named Demis Hassabis is reading Gödel, Escher, Bach and concluding, before most of his professors would have agreed, that first-order logic can't be the full answer to building intelligence. Sebastian Mallaby spent years inside that story. His new book, The Infinity Machine: Demis Hassabis, DeepMind, and the Quest for Superintelligence, is the most serious attempt yet to explain not just what AI is, but why the people building it can't stop. His answer draws on a line Jeff Hinton borrowed from Robert Oppenheimer: invention is sweet. A scientist, given the chance to build something, simply cannot resist. The consequences come later. In this conversation, Mallaby joins Josh Tyson and Robb Wilson to explore the full sweep of the Demis Hassabis story — from game designer to neuroscientist to Nobel laureate to the man running Google's flagship AI lab. They talk about why DeepMind was built the way it was, with neuroscientists and physicists and probabilistic mathematicians before AI was even a field, and how that cross-disciplinary foundation ended up mattering more than anyone expected. They talk about what the defeat of the world Go champion felt like from the inside, the humans who gave up and the ones who discovered new depths. And they talk about what it means that the internet, a thing nobody built to train AI, turns out to be exactly the fuel the industrial revolution of intelligence needed. Demis's own metaphor: it's like dinosaurs that died and turned into oil. Nobody designed it for this. It just happened to be there. The conversation also gets into what Mallaby calls the infinity machine: the reason the kind of inductive learning AI uses requires almost infinite examples to be reliable, and why the name captures something the scaling law charts obscure. Why the internet taught us more about the range of human experience than Hassabis expected. Why gaming runs so deep through the entire history of machine intelligence. And what it actually means to ask whether a machine is intelligent, when the people who built DeepMind weren't sure they had a definition. ---------- Support our show by supporting our sponsors! This episode is supported by OneReach.ai Forged over a decade of R&D and proven in 10,000+ deployments, OneReach.ai’s GSX is the first complete AI agent runtime environment (circa 2019) — a hardened AI agent architecture for enterprise control and scale. Backed by UC Berkeley, recognized by Gartner, and trusted across highly regulated industries, including healthcare, finance, government and telecommunications. A complete system for accelerating AI adoption — design, train, test, deploy, monitor, and orchestrate neurosymbolic applications (agents). - Use any AI models - Build and deploy intelligent agents fast - Create guardrails for organizational alignment - Enterprise-grade security and governance Book a free demo: https://onereach.ai/book-a-demo/?utm_source=soundcloud&utm_medium=social&utm_campaign=podcast_s7e6&utm_content=1 ---------- The revised and significantly updated second edition of our bestselling book about succeeding with AI agents, Age of Invisible Machines, is available everywhere: Amazon — https://bit.ly/4hwX0a5 #ai #invisiblemachines #podcast #techpodcast #aipodcast #deepmind #DemisHassabis #InfinityMachine #agi #machinelearning #alphago #futureofai

The IRS has roughly 60,000 fax machines, and nobody can get rid of them. Not because there’s a law that says you have to use them (there almost certainly isn’t), but because likely decades ago a memo got written, somebody interpreted fax machines as the most secure transmission method, and that memo calcified into what Jennifer Pahlka calls "folk law," a perceived rule that nobody can locate, nobody can challenge, and everybody treats as immutable. Folk law looms large in the American government right now. Cascades of rigidity built from outdated interpretations of rules that were flexible to begin with, administered by people who were never asked whether any of it was working. Jennifer Pahlka, who wrote Recoding America: Why Government Is Failing in the Digital Age and How We Can Do Better, is the founder and former executive director of Code for America, and was Deputy CTO for Government Innovation in the Obama White House. She’s working on the gap between what government is supposed to do and what it actually does. In this conversation, Robb, Josh, and Jennifer go deep on what’s actually broken and what it would take to fix it. The folk law problem is real, but it's not the deepest one. The deeper dysfunction: government is structurally designed to be faithful to process rather than outcomes. Oversight bodies don't ask whether people got the benefit. They ask whether you followed the procedure. That incentive structure produces "rationing by friction" — where the hardest programs to navigate self-select for the people who need help least and exclude the people with the most chaotic lives, the fewest resources, and the most at stake. Her Recoding America team is already working with states to build something Robb describes as a P&L for regulation. Not just removing rules, but assigning friction costs, finding where wet signatures are still required for no reason, and surfacing the trade-offs that have never been explicitly named. LLMs are uniquely good at this. The question isn't whether the technology can help. It's whether the political will to use it correctly can be assembled in time. ---------- Support our show by supporting our sponsors! This episode is supported by OneReach.ai Forged over a decade of R&D and proven in 10,000+ deployments, OneReach.ai’s GSX is the first complete AI agent runtime environment (circa 2019) — a hardened AI agent architecture for enterprise control and scale. Backed by UC Berkeley, recognized by Gartner, and trusted across highly regulated industries, including healthcare, finance, government and telecommunications. A complete system for accelerating AI adoption — design, train, test, deploy, monitor, and orchestrate neurosymbolic applications (agents). Use any AI models - Build and deploy intelligent agents fast - Create guardrails for organizational alignment - Enterprise-grade security and governance Book a free demo: https://onereach.ai/book-a-demo/?utm_source=soundcloud&utm_medium=social&utm_campaign=podcast_s7e5&utm_content=1  ---------- The revised and significantly updated second edition of our bestselling book about succeeding with AI agents, Age of Invisible Machines, is available everywhere: Amazon — https://bit.ly/4hwX0a5 #ai #government #govtech #JenniferPahlka #RecodingAmerica #publicpolicy #enterpriseai #doge #bureaucracy #invisiblemachines #podcast #techpodcast #aipodcast