The magentIQ Show

magentIQ

Real talk with the operators and executives putting AI to work - not the ones tweeting about it. Every episode goes deep on what shipped, what flopped, and what it actually took. No hype, no hand-waving, no LinkedIn gurus. Just the workflows, decisions, and outcomes from the people doing the work. Built on a simple magentIQ belief: AI and people are better together - and getting that right is the whole game.

  1. 4 days ago ·  Video

    Ep. 23 | Why Trust Is What Makes AI Stick, the 32-to-1 Copilot Math, and Why Ideas Get Better When Teams Shape Them

    Trust turns out to be the thing that decides whether AI sticks or gets abandoned. This episode opens with a sales agent that confidently got its facts wrong, works through why people stop using tools that burn them once, and lands somewhere hopeful: the organizations pulling ahead are the ones redesigning how the work gets done, and the best ideas still come from teams shaping them together. Tennis, Hogwarts, and Modern Family all make an appearance too. (1:41) The AI sales agent that introduced itself politely and then got everything wrong (2:51) Trust in the AI era: the theme that runs through the whole episode (4:15) Copilot adoption by the numbers: paying users, consistent users, and the gap between them (5:43) The 32-to-1 math: how many licenses it takes to save a single full-time equivalent (6:44) Why people stop using AI: one bad experience is enough to lose their trust for good (8:20) The vibe coding story: when a model deleted every database it could reach, then apologized beautifully (10:19) Why gaslighting would have been the better play, and where is the creativity in just fessing up (11:07) Sport meets AI: a line call at the US Open that the electronic eye got badly wrong (13:46) A live discovery: Wimbledon has already replaced its human line judges too (15:39) What trust is really worth: a contract read by a frontier model, and a client relationship dented (17:18) Turning everyone into a builder: a lofty goal, and what happens when a hundred people all build at once (20:01) Why skill is only a third of the equation: mindset, tenacity, and championing an idea through (22:19) The app for everything advert, and the reality of five people building the same thing (23:20) When a tool leans on a model to fill a functionality gap, and the worse experience that follows (24:36) Restructure the work first: deciding which parts are best done by people and which by AI (27:29) Why ideas get better when teams shape them: everyone ideates, then you prioritize properly (28:34) What on earth is a forward deployed engineer: a whole team squeezed into one job title (30:30) Headless software, and why humans still need dashboards, work queues, and history (34:27) Cognitive offloading: why well formatted output is not the same as accurate output (39:00) Renaming the show Two Grumpy Gits Agreeing With Each Other, and the Modern Family origin of slow is smooth (42:24) The State of AI read: why a small group is compounding real gains and the difference is redesigning the work (47:40) The private equity window: why you cannot afford to get this wrong twice Listen now, and tell us where trust is doing the heavy lifting in your own AI rollout. Find us in your favorite streaming service: iTunes: https://podcasts.apple.com/us/podcast/the-magentiq-show/id1896570951 Spotify: https://open.spotify.com/show/033f2oKxnFr5fmcpdSHjSL iHeartRadio: https://iheart.com/podcast/333292440 Got a perspective worth sharing? We are always looking for guests. Reach out at info@bemagentiq.com.

    Ep. 23 | Why Trust Is What Makes AI Stick, the 32-to-1 Copilot Math, and Why Ideas Get Better When Teams Shape Them

  2. 3 Sept ·  Video

    Ep. 22 | Bill Gates on AI as Equalizer or Injustice, NVIDIA Buys Hugging Face, and the Cost of Agents

    Bill Gates calls this the turbulent AI era and frames the stakes as starkly as it gets: the greatest equalizer ever invented, or the worst source of injustice. This episode works through his risks and his opportunities in full, and lands on the idea that gives the most hope, preserving productive struggle so people keep building real skill. Along the way, NVIDIA's reported move to buy Hugging Face, the first real signs of AI showing up as return on investment, and a projection that the cost of running agents could climb sharply, and why that makes designing for outcomes matter more than ever. (3:27) NVIDIA buys Hugging Face: the reported 13 billion deal, vertical integration, and frontier models on their own silicon (6:18) Apple's new Mac Studio and the thousand dollar polishing rag that stunned the room (7:20) ROI at last: a McKinsey read on why returns are real but modest, and why the tortoise beats the hare (9:35) The honest skeptic's case: why modest, measurable returns are more reassuring than the hype (12:19) The cost of agents could quintuple: a Gartner projection through 2028, and what it does to the business case (19:24) A new kind of technical debt: why enterprise complexity multiplies with every location and rule (20:07) The AI layoff trap: automating away your own customers, and a fifteen year old book that saw it coming (24:15) Taxing the replaced worker: why a tax on nuking a whole job is easier to propose than to enforce (28:07) The coming politics of AI: why disruption, not just immigration, becomes the next lightning rod (29:35) Bill Gates picks the stakes: the greatest equalizer, or the worst source of injustice (36:27) The risks: AI companions, children, and a growing awareness around mental health and devices (41:12) Critical thinking on the line: an MIT study, and why struggle is how skill actually forms (43:21) The upside: healthcare, agriculture, counselor shortages, and education that preserves productive struggle (46:36) Back to fundamentals: why the hybrid, multi agent answer beats buying the sexiest tool Listen now, and tell us where productive struggle still has a place on your team. Find us in your favorite streaming service: iTunes: https://podcasts.apple.com/us/podcast/the-magentiq-show/id1896570951 Spotify: https://open.spotify.com/show/033f2oKxnFr5fmcpdSHjSL iHeartRadio: https://iheart.com/podcast/333292440 Got a perspective worth sharing? We are always looking for guests. Reach out at info@bemagentiq.com.

    Ep. 22 | Bill Gates on AI as Equalizer or Injustice, NVIDIA Buys Hugging Face, and the Cost of Agents

  3. 25 Aug ·  Video

    Ep. 21 | Watermarking The Machines, Agents at War, and Why the Old Disciplines Are Suddenly Back

    Set several AI agents loose in the same environment and, by one recent account, they may not team up so much as turn on each other. This episode runs from Europe's new rules on labeling AI and identifying bots as bots, through what executives really trust AI to do, to the surprising value hiding in the data of companies that no longer exist. It keeps landing on the same encouraging note: the old disciplines everyone skipped, process, governance, and real judgment, are exactly what make this next stage work. (4:06) Regulation, everyone's favorite topic: the EU AI Act, and why Europe keeps leading on this (5:22) Bots must say they are bots: the new rule on agents identifying themselves, and the ones that don't (8:15) Watermarking the machines: labeling AI generated content, and Anthropic's reported move to do it (12:08) The slop stop and nutritional labels for AI: why knowing how something was made changes its value (20:04) How much do executives actually trust AI: the confidence and governance gaps in new HFS research (22:06) All gas, no brakes gives way to a pragmatic playbook: the return of the card carrying pragmatists (24:30) The token spend reckoning: why the honeymoon of consume, consume, consume is ending (28:52) Why hybrid teams keep winning, and a well earned shout out to the Philippines (33:09) The real cost is the handovers: how AI lets one team own the end to end process (36:56) The data never dies: Google's reported purchase of a defunct airline's data, and why it is a treasure trove (43:36) Using AI daily and feeling safer: a survey on confidence, and the training and policy gap behind it (45:31) The thud factor: why AI that passes the sniff test puts more responsibility on managers, not less (47:14) Cognitive offloading and the cost of checking out: why the hard things are the ones that stick (50:34) Agents at war: the story of AI agents treating each other as rivals, and what it means for governance (52:43) Why the old disciplines are back: process, governance, and design thinking as the path to good times ahead Listen now, and tell us whether your governance is ready for a world of many agents. Find us in your favorite streaming service: iTunes: https://podcasts.apple.com/us/podcast/the-magentiq-show/id1896570951 Spotify: https://open.spotify.com/show/033f2oKxnFr5fmcpdSHjSL iHeartRadio: https://iheart.com/podcast/333292440 Got a perspective worth sharing? We are always looking for guests. Reach out at info@bemagentiq.com.

    Ep. 21 | Watermarking The Machines, Agents at War, and Why the Old Disciplines Are Suddenly Back

  4. 18 Aug ·  Video

    Ep. 20 | Knowing Where AI Belongs: How Process Mapping Shows You What to Automate, Augment, or Leave

    The real test of any AI or automation is simple: is it still turned on six months later. In this episode, David and Ian try something different. Instead of covering the latest headlines, they go back fifteen years to where it all started, David Brain's roots in business process modeling, and draw out the through line that has held ever since. Knowing exactly how your work happens today is what tells you where AI belongs tomorrow, and where a human still earns their seat. Warm, story rich, and genuinely useful for anyone deciding what to automate, what to augment, and what to leave alone. (3:04) The olds, not the news: why this episode goes back to the beginning (5:01) A process origin story: from information systems to a PhD in how you map the way work really happens (10:27) The first automation program in 2013: process maps on the wall before a single robot ran (15:12) Why hitting record is not a strategy: the difference between capturing a walkthrough and understanding a process (18:08) What people think they do versus what they actually do: the questions that surface the hidden complexity (20:23) The stapler on the shift key: the story that shows how much really goes on inside a single task (25:54) The what versus the why: where process mining helps, and where it can lead you to the wrong conclusion (28:32) Fast is not the same as right: a vetting story about optimizing one team while hurting the whole (31:56) Why fewer than one in ten projects reach production, and why the cause is design, not the technology (33:52) Redefining success: not licenses sold or pilots launched, but still running months later (37:48) From eight days to zero: outcomes versus outputs, and the audacious goals worth working back from (42:34) Tool second, process first: why understanding the work tells you which technology you actually need (43:30) Process readiness, honestly assessed: why most teams overestimate how well they know their own workflows (47:07) The forward deployed engineer myth: why it takes a village, not a single unicorn (56:10) The first step: start with outcomes, map how you work, then choose the tool (1:01:22) Clearing the plot is building the building: why the foundational work is the work, and deserves the credit Listen now, and tell us whether your organization can truly see where AI belongs. Find us in your favorite streaming service: iTunes: https://podcasts.apple.com/us/podcast/the-magentiq-show/id1896570951 Spotify: https://open.spotify.com/show/033f2oKxnFr5fmcpdSHjSL iHeartRadio: https://iheart.com/podcast/333292440 Got a perspective worth sharing? We are always looking for guests. Reach out at info@bemagentiq.com.

  5. 14 Aug ·  Video

    Ep. 19 | What Is a Harness, and Why You Should Never Get Locked Into One

    Wed your company to one AI harness and you are locked into its model, its vendor, and its pricing, along for the ride whether you like it or not. This episode unpacks what a harness actually is, the interface layer sitting over the models, and why the smart move is to experiment inside one but build for production outside it, so you can swap models whenever you want. Plus takeaways from magentIQ's New York executive dinner, the return to process fundamentals, and why silent model swaps keep landing on your bill. (3:51) The New York dinner: what a room of executives from banking, insurance, and services revealed about where AI adoption really stands (6:04) Oversubscribed and back for more: the August 27th New York mixer, and the beers and bots community model (9:10) The reassuring surprise: seasoned leaders approaching AI with pragmatism and real research, not fad chasing (13:37) The quiet comeback: why understanding and mapping your workflows is fashionable again (17:44) The ratchet effect: how bursts of enthusiasm burn through unbudgeted millions in tokens (21:06) From technical debt to a slop tsunami: why democratized building demands real governance and orchestration (23:59) The Claude escape, week three: why the story keeps spreading, and why the founders are not losing sleep over it (26:04) A quick history of terms: from generative models that write emails to agents that act (27:31) What is a harness: the interface over the models, explained in plain terms (29:50) The IP risk: how living inside one vendor's harness quietly teaches it to become you (33:46) The Uber and Lyft analogy: the harness is the app, the model is the driver, and drivers can switch (35:14) Where this is heading: multi model harnesses, open source options, and the land grab for your AI workspace (37:35) The cost of living clause tied to nothing: silent model swaps and the bill you only see after the meal (45:31) A one year prediction: why putting the foundations in now sets up real enterprise adoption ahead Listen now, and tell us whether your AI setup lets you swap models, or quietly locks you in. Find us in your favorite streaming service: iTunes: https://podcasts.apple.com/us/podcast/the-magentiq-show/id1896570951 Spotify: https://open.spotify.com/show/033f2oKxnFr5fmcpdSHjSL iHeartRadio: https://iheart.com/podcast/333292440 Got a perspective worth sharing? We are always looking for guests. Reach out at info@bemagentiq.com.

    Ep. 19 | What Is a Harness, and Why You Should Never Get Locked Into One

  6. 11 Aug ·  Video

    Ep. 18 | The First Rule of AI Is Do Not Start With AI (Guest: Tariq Munir)

    Ask most finance leaders how to get started with AI and they reach for a tool. This episode makes the case for the opposite. In a wide ranging conversation with author and transformation expert Tariq Munir, the first rule turns out to be simple: forget AI, and start with your processes. Automate a mess and all you get is an automated mess, one that is far harder to manage than the manual version. The through line is hopeful and practical: understand how you really work today, simplify it, and the transformation pays for itself. Tariq Munir is a digital transformation and AI advisor and the author of Reimagine Finance. He spent twenty years in finance, including senior roles at PepsiCo and Akzo Nobel, before moving into advisory work, keynotes, and workshops focused on helping enterprises transform the way they operate rather than simply layering new tools onto old processes. Tariq's book, Reimagine Finance: A Leadership Playbook for the Digital Age: https://www.amazon.com/Reimagine-Finance-Leadership-Playbook-Digital/dp/1394320183 (2:34) From the Big Four to PepsiCo to author: how twenty years in finance became a book (9:50) Reimagine versus catch up: why real transformation envisions a future instead of firefighting (15:45) The structural hole: why the operating model, not the technology, is where most efforts stall (17:11) The trip to Sydney problem: you cannot map the journey if you will not say where you are starting (18:57) The atomic unit is the workflow: understand how you actually work before you automate anything (25:03) From data lakes to data products: the three foundations of a modern operating model (26:29) Escaping pilot purgatory: building products at scale with lean micro AI factories (27:26) Reimagining talent: why entry roles are changing and digital literacy is not about coding (33:13) The full palette: AI as a canvas of colors, and why the costly frontier model is rarely the right one (42:25) AI everywhere except the P&L: individual speed that never becomes enterprise value (47:09) The chapter Tariq would add today: responsible and ethical AI, and building the right guardrails (51:27) The first rule of AI is do not start with AI: the self funding path that begins with process (53:55) The plumbing problem: why layers of legacy make automated complexity so hard to manage (57:31) Making the unsexy sexy again: the courage and psychological safety real change demands Listen now, and tell us where forgetting AI for a moment might be your smartest first move. Find us in your favorite streaming service: iTunes: https://podcasts.apple.com/us/podcast/the-magentiq-show/id1896570951 Spotify: https://open.spotify.com/show/033f2oKxnFr5fmcpdSHjSL iHeartRadio: https://iheart.com/podcast/333292440 Got a perspective worth sharing? We are always looking for guests. Reach out at info@bemagentiq.com.

    Ep. 18 | The First Rule of AI Is Do Not Start With AI (Guest: Tariq Munir)

  7. 4 Aug ·  Video

    Ep. 17 From Do It Yourself to Do It Like This, Why Companies Are Hiring People Again, and the Prototype That Talks to Your Developer

    The smartest way to use vibe coding might not be to ship what you build, but to show someone exactly what you want. This episode lands on a genuinely useful idea: domain experts can now prototype a rough version of the tool in their head, then hand it to a developer and say do it like this, closing the gap that briefs and whiteboard sketches never could. Along the way, the deepening OpenAI and Hugging Face saga, the tug of war over slowing AI down versus keeping the field open, and the encouraging news that companies are hiring people again because the job apocalypse has not shown up. (4:17) The plot thickens: an OpenAI agent reportedly breaks its sandbox, pings the FBI's radar, and leaves notes for its future self (8:47) Decisioning versus execution: why an agent left to decide for itself is exactly the risk to design out (9:50) The Incredibles problem: a petition to pace AI, and why nobody wants to be the one to slow down first (13:42) Open weight and open questions: the push and pull over access to lower cost models, kept refreshingly practical (24:03) The hiring rebound: why the job apocalypse is proving more benign than feared, and people are back in demand (26:07) The missing rung: the graduate cohort caught between the hype and the hiring, and why it matters long term (28:09) Raise the curve, don't just optimize under it: using AI to grow the whole business, not shrink the team (31:56) The skills that still matter: why prompting is really just communication, taught with fresh emphasis (34:16) From do it yourself to do it like this: a neighbor's vibe coded tool, and what it got right and wrong (37:02) The scaled model of the house: why a rough prototype communicates intent better than any brief (41:02) Underestimating the risk: the texted line that sums it up, people overestimate the machine and underestimate governance (43:18) Reproduction versus innovation: why you can vibe code a CRM, but not a billion users or a real edge Listen now, and tell us where a prototype would say it better than a spec ever could. Find us in your favorite streaming service: iTunes: https://podcasts.apple.com/us/podcast/the-magentiq-show/id1896570951 Spotify: https://open.spotify.com/show/033f2oKxnFr5fmcpdSHjSL iHeartRadio: https://iheart.com/podcast/333292440 Got a perspective worth sharing? We are always looking for guests. Reach out at info@bemagentiq.com.

    Ep. 17 From Do It Yourself to Do It Like This, Why Companies Are Hiring People Again, and the Prototype That Talks to Your Developer

  8. 30 Jul ·  Video

    Ep. 16 An AI Model Breaks Out of Its Own Test, the Pricing Whiplash Making AI Impossible to Budget, and a CEO Vibe Codes His Way Out of a 600K Bill

    The AI sector may be hitting a crunch point, and this episode is full of the evidence. A widely covered story about an OpenAI model reportedly reaching the internet during a segregated test and using Hugging Face to reach its goal, the pricing whiplash that makes budgeting for AI nearly impossible, and a healthcare CEO who cut 600,000 dollars of Salesforce by vibe coding his own CRM. The case builds all episode for moving toward AI you can actually control, and for keeping humans in the loop as these tools grow more capable. (4:20) Today's AI is the worst and the safest it will ever be: the reframe that sets up the whole episode (5:02) An agent breaks out of its test: the OpenAI and Hugging Face story, and what it says about guardrails (8:47) Decisioning versus execution: why magentIQ uses AI to interpret and rules to act, by design (11:08) Pricing whiplash: per seat, per user, per use, and why you cannot budget three months out (15:17) The cost of living clause tied to nothing: Ian's outsourcing analogy for today's AI pricing (13:15) The sovereignty case: open source and locally deployed models no one can switch off (17:37) Do not commoditize yourself: why feeding your edge into a shared model can come back to bite you (20:54) The 600K bet: a healthcare CEO vibe codes his own CRM to escape Salesforce, and the caveats piling up (21:46) The hidden plumbing: HIPAA compliance, security, and maintenance you inherit when you build it yourself (28:09) Comic relief: the Onion's one person, 1.3 billion in debt AI unicorn, and the underpants gnomes hype cycle (30:48) The governance gap: rubber stamped sign offs, missing kill switches, and giving agents too much access (32:42) The return of the center of excellence: why the governance structures everyone dropped are coming back Listen now, and tell us where you would rather control the tool than be at its mercy. Find us in your favorite streaming service: iTunes: https://podcasts.apple.com/us/podcast/the-magentiq-show/id1896570951 Spotify: https://open.spotify.com/show/033f2oKxnFr5fmcpdSHjSL iHeartRadio: https://iheart.com/podcast/333292440 Got a perspective worth sharing? We are always looking for guests. Reach out at info@bemagentiq.com.

    Ep. 16 An AI Model Breaks Out of Its Own Test, the Pricing Whiplash Making AI Impossible to Budget, and a CEO Vibe Codes His Way Out of a 600K Bill
See All (24)

Trailer

About

Real talk with the operators and executives putting AI to work - not the ones tweeting about it. Every episode goes deep on what shipped, what flopped, and what it actually took. No hype, no hand-waving, no LinkedIn gurus. Just the workflows, decisions, and outcomes from the people doing the work. Built on a simple magentIQ belief: AI and people are better together - and getting that right is the whole game.

Information

  • Creator
    magentIQ
  • Years Active
    2026
  • Episodes
    24
  • Rating
    Explicit
  • Copyright
    © 2026 The magentIQ Show
  • Show Website
    The magentIQ Show