Unscripted with Jeff Pedowitz

Your best employee might be the reason you can’t scale.
I’ve hired them, and I’ve defended them in meetings. There was the account manager who crushed his number every year and was universally disliked by everyone around him. There was the consultant who was brilliant with clients and quietly toxic to the team.
It took me too long to see it. You can put your values on every wall in the office, but if you keep employing people who break them, your team hears a very different message.
This week on Unscripted, I sat down with Russ Reeder, who has put a name on this problem.
Russ has led nine PE-backed companies across six continents. He started Rightsline, which now runs the rights and licensing back office behind Netflix, Hulu, Sony and Universal. He helped sell Media Temple to GoDaddy. Today he runs KeyDelta, helping leadership teams fix the operating model before they bolt AI onto it.
His new book is called Heroes Don’t Scale.
A “hero” is the engineer who is the only one who knows how the code works. It’s the rep who hoards every client relationship, or the manager who treats information like currency. Everyone celebrates them, and the company slows down around them.
Russ has replaced five CEOs. Three of them were heroes.
A few more moments from the conversation stuck with me:
Back in 2006, his team had encrypted movies streaming from hotel basements into people’s homes. They had the studio deals and the set-top boxes, and Warner Brothers was ecstatic. One controlling shareholder killed it. You know who built that business instead.
AI is about to create a new generation of heroes. Picture the one employee who vibe-coded your sales automation and is the only person with the login.
People who are scared of AI taking their job should be scared. Russ argues the cure is curiosity, because you can’t be curious and afraid at the same time.
Companies are bought, not sold. Cutting your way to $5M more in EBITDA doesn’t fool the buyer.
Russ asks one test question in his book: if your key person quit tomorrow, would the business still work?
Most leaders say no. Russ says nearly every company would survive. The fear of losing that person is usually the real problem.
So tell me in the comments: who is the hero in your company, and what is it costing you?
 
#Leadership #AI #PrivateEquity #Scaling #Unscripted

Most economic forecasts come from people with something to sell. Erik Britton is a rare exception.
He began his career inside the Bank of England, went on to help lead Oxford Economics, and now runs Fathom Consulting in London, one of the few genuinely independent macro shops in the world, where governments, global asset managers, and pension funds go for analysis that sits where economics, geopolitics, and markets meet.
In this conversation, Erik makes a call most people in the AI conversation won’t make out loud: either this is the most transformational technology in human history, bigger than electrification, the computer, or the jet engine, or it’s a bubble. He takes the bubble side, and walks through the arithmetic behind it.
We get into:
Why current AI capex implies returns that would require a 10 to 15 percent increase in US GDP
The difference between a dot-com bubble and a 2008 bubble, and why hidden debt in shadow banking, private equity, and venture capital makes this one closer to the second
Why AI displacement is hitting the top of the skills distribution instead of the bottom, and what that does to the apprenticeship path into senior roles
How a central bank should classify AI: positive supply shock, negative demand shock, and financial stability risk, all at once
How Erik actually uses AI in his own work, and why he hasn’t embedded it in Fathom’s analytical workflow
What China’s dual circulation strategy is really optimizing for, and why it isn’t making its own people rich
Why every forecast critic is already running a forecast, whether they admit it or not
His direct advice for leaders building a 2027 plan: expect to get punched in the guts, and start anyway
A conversation for CEOs, CFOs, and revenue leaders who have to commit to a number in an environment nobody can model cleanly.
Read the companion article: https://www.pedowitzgroup.com/blog/its-a-bubble-or-its-the-biggest-thing-that-ever-happened-erik-britton-on-planning-for-2027
Learn more about Fathom Consulting: https://www.fathom-consulting.com
#Unscripted #AIBubble #MacroEconomics #B2B #RevenueMarketing #AIStrategy #ThePedowitzGroup

"I've been to space. I've seen Earth from the outside.
And I'm more worried about what's happening down here than anything I saw up there."
Colonel Terry Virts has flown 40 combat missions, piloted the Space Shuttle, commanded the International Space Station, directed a documentary, and earlier this year ran for Congress in Texas.
He came on Unscripted to talk about AI. We ended up talking about all of it.
The thing that stayed with me: he teaches a case study at Harvard Business School about the Challenger and Columbia disasters. Engineers were warning NASA. Managers had schedule pressure. Warning signs were normalized. Seven people died.
His TED Talk argument: the AI moment looks exactly the same.
Except this time, the carrot isn't a launch date. It's a trillion-dollar IPO. And the consequences of getting it wrong aren't seven lives.
A few things from this conversation I can't stop thinking about:
"Don't normalize deviance. If something bad's happening and it's not killing you, don't assume it's fine. Maybe you're just getting lucky."
The night he and a Russian cosmonaut watched bombs explode over Ukraine from the ISS window — and didn't say a word
He sold his Russian space suit and donated the money to Ukraine
His $5-a-day government per diem after 200 days in space
Why bioweapons, not chatbots, are what actually keep him up at night
One of the most wide-ranging conversations I've had on this show. Space, AI, democracy, corruption, civic duty — and why the founding fathers anticipated Trump but never anticipated Mike Johnson.
Full episode is live now. Link in the comments.
#AI #Leadership #Space #Democracy #Unscripted

Charlotte Duncan has been teaching AI since 2018 — before most people knew they should care. She wrote the first AI course in North Carolina. She helped build the first national K-12 AI guidelines. She trained at MIT. And then Mark Cuban called.
As Chief Learning Officer at the Mark Cuban Foundation, Charlotte runs free AI boot camps for high school students across all 50 states. Not the kids whose parents already work in tech. The rural kids, the first-generation kids, the students from Boys Town in Omaha who come in carrying real trauma and leave building apps. Three Saturdays, 20 hours, one problem they actually care about in their own community — and nobody is allowed to make a PowerPoint.
Her worry isn't that AI is too powerful. It's whether the kids coming up will have any real say over it at all. Right now, only 12% of American high school students take a computer science class in a given year. The standards that govern what gets taught take up to a decade to update. And the AI cycle moves every few months.
We get into what it actually looks like to teach a generation that's afraid of a technology they're already using, why she thinks teenagers are more ready to fix this than the adults in charge, and what she told me at the end that I genuinely didn't see coming: put down the AI tools for a minute and read a book.
Charlotte Duncan, Unscripted. Let's get into it.

Nico Ianeselli wanted to be a doctor like his father until he heard it took 11 years, so he earned dual master's degrees in engineering and architecture instead. In this conversation with Jeff Pedowitz, the Chen Moore and Associates technology chief explains why construction's slowness isn't resistance but survival (a missed error means collapse, not a bug fix), and why safety-critical industries need humans in the lead, not just in the loop. He breaks down what digital twins actually do, from simulating hurricanes and full-load bridge scenarios to the 4D construction sequencing that avoided major mistakes on the $2.2 billion Muskrat Falls dam, and makes the case that IoT sensors plus digital twins could stretch a bridge's 50-year lifespan toward 200. Fresh from a conference trip to China, he delivers a blunt geopolitical read: Europe is crawling, America is walking, China is running. He takes on the unanswered AI liability question hanging over the Professional Engineer's stamp, flips the data center energy debate with his 4% versus 96% argument, and closes with a 50-year vision of AI contributing genuinely new engineering principles. His advice to a fearful Gen Z: magic is on the other side of fear.

Jennifer Preston is a biomedical engineer with a PhD who now leads AI strategy for the consultants shaping drug development and commercialization decisions across pharma and biotech. Her firm uses AI to create synthetic patients and synthetic physicians, digital stand-ins that get interviewed and surveyed to test drug messaging before any real human is involved. She joins Jeff to answer the uncomfortable question underneath it all: what does it mean when healthcare starts making decisions based on people who don't exist?

Matthew Guggemos spent more than 20 years as a speech-language pathologist teaching children to communicate — kids for whom language doesn't come free — and now builds AI at scale as CTO of iTherapy, the company behind InnerVoice, an AAC app with over 50,000 users. He's also a conservatory-trained drummer who played in Michigan State's top jazz band, and he joins Jeff to explain why music, language, and AI are all the same problem: getting nearly infinite expression out of a finite set of constraints.

Camberley Bates has spent 30-plus years as one of enterprise technology's most respected independent analysts, paid to cut through vendor hype and tell executives what's real. On this episode of Unscripted, she and Jeff unpack what IBM's earnings dip really signals (AI infrastructure spending delayed mainframe upgrades, not weakness), why 70 to 80 percent of enterprise data still lives on-prem, and why cloud-first has given way to workload placement. She explains the physics nobody can outrun: latency and data gravity make moon-based data centers a fantasy and on-prem AI stacks a boom for Dell and HPE. Bates gets candid about prompt privacy (she wouldn't bet enterprise data stays out of training models), the SaaS valuation crisis, the data center NIMBY wars, and how to decode "AI-ready" marketing. Her prediction for the claim that ages worst: that AI eliminates work. People need work to have value in their souls.

Russ Wilcox, CEO of ArtifexAI, went from a broken sewer decision in his Cape Cod hometown to advising the Pentagon and State Department on the US-China AI race. In this episode of Unscripted, he tells Jeff how mining boring public records, zoning codes, meeting minutes, and permit archives, unlocks the 85 percent of human knowledge trapped in unstructured data. He explains why he believes the US is currently losing the AI race: China treats AI as infrastructure, with data reservoirs and specialized swarm agents, while the West chases god models built on 15 percent of human knowledge. Wilcox maps China's chip packaging dominance, makes the case for cognitive sovereignty, and lays out his three-part playbook for bringing AI to local government: compliance, decision support, and transparency. His vision: AI that revives local democracy through public-private partnership instead of digital techno-feudalism.

Every impressive AI demo sits on an enormous, expensive, and very unglamorous foundation of data. Keith Pijanowski is one of the people who built it. After a decade-plus at Microsoft (where he helped evangelize Azure before its public release) and data pipeline work at BNY Mellon, he now works on the storage infrastructure that AI training and inference actually depend on. He breaks down why hundreds of GPUs hammering storage for months makes training a workload the software industry has never seen, and why inference is no longer lightweight thanks to KV caching across GPU memory, system memory, and network storage. He predicts power ratings will soon rival performance ratings for GPUs, dismisses space data centers as science fiction while taking ocean data centers seriously, and argues the build-out is rational because the Jevons paradox guarantees efficiency gains produce more demand, not less. His deepest concern isn't the machines. It's people publishing content and shipping code they don't understand. Use AI as a productivity tool, but don't let it turn you into an operator.