What Survived the AI Hype: Lessons from 2035

Olivia Carter

Hey everyone, welcome back. I’m Olivia Carter, here as always with Mark Putnam. Today we’re time traveling—sorta. We’re jumping to Timothy Chester’s vision of 2035, where we ask: after all the AI mania and grand promises, what really lasted? Mark, this is right in your wheelhouse, isn’t it?

Mark Putnam

It is. And honestly, Liv, this isn’t my first dance with a technology bubble. I remember the dot-com bust, clear as day—Texas A&M, late nineties. Everyone in tech was pushing major web projects, and then suddenly... the bottom fell out. Budgets frozen. Fear everywhere. But you know what? That’s when the actual durable work got done. The hype faded, and a kind of discipline returned. The leaders who surfaced then? They stopped spinning tales and focused on what was real—work simplification, rethinking, slow, steady progress.

Olivia Carter

It sounds—well, a little too familiar. I mean, during the AI boom of the 2020s, every CEO had to have some “AI transformation office,” right? It was all about urgency, speed, and... sometimes just trying to look clever in front of a boardroom. Actual infrastructure? Sort of an afterthought.

Mark Putnam

Exactly. Speed became the proxy for leadership—move fast, be the serious one, don’t let anyone call you cautious. But, as with the internet before, vision raced ahead of actual institutional readiness. The reality? The technology didn’t fail—the economics did. When the funding dried up, reality set in. Not collapse, just... clarification. That’s what people forget—real transformation didn’t happen during the pop. It came after.

Olivia Carter

Yeah, and for a lot of organizations, that correction was the only way to get honest about what was working versus what was hype. There’s a throughline here—every major tech runs this cycle. Exuberance, overreach, and then, eventually, actual maturity. And if you aren’t ready for that reset point, you just don’t make it to the “real” part.

Olivia Carter

So let’s land in 2035. The AI gold rush is behind us. What’s left standing? Tim Chester nails this in his vision: AI isn’t a headline anymore. It’s like, well, plumbing—quiet, everywhere, completely taken for granted. I know for us in journalism back then, it started with these shiny, overhyped reporting tools, all these dashboard-y outputs. But honestly, by the time the dust settled, what we valued was totally different.

Mark Putnam

That resonates. By 2035, nobody’s hiring “prompt engineers.” Drafting, summarizing, even complex pattern recognition—those all became table stakes, not differentiators. The real constraint became judgment. The folks who advanced in their careers weren’t the fastest at executing—they were the clearest thinkers, the ones you could trust to take responsibility for the output. I keep saying: AI amplifies judgment, not replaces it. You still need people who know how to ask the right questions and challenge the answers.

Olivia Carter

Mark, it took ages for newsrooms to figure that out. After the AI razzle-dazzle, my colleagues at Wired started spending way less time just churning out pieces, and way more on shaping stories, interviewing, fact-checking—actual editorial judgment. It was almost like AI handled the noisy stuff so we could be more... human. But, to be clear, the skills that mattered most post-boom were never automation tricks. It was that old-school ability to interpret and own the final call.

Mark Putnam

It’s the same across higher ed. The competitive advantage quietly shifted from “how fast can you automate X?” to “who can make the right call when the machine output is ambiguous or—worse—completely wrong?”

Mark Putnam

Let’s talk about how AI actually got woven into places that matter. Early on, most AI tools sat on the margins—little portals here and there, ask the bot, export the spreadsheet. Nobody truly changed process, so adoption fizzled. It wasn’t until the early 2030s that we saw the shift: AI embedding itself right into ERP, CRM, finance, HR—the beating heart of university operations.

Olivia Carter

Yeah, everywhere else too. I always joke about those standalone “AI for reporters” products I demoed. Flashy, but the minute you had to jump platforms and change how you worked, people bailed. But when that stuff just became part of the platform—like, summarizing or tagging as you wrote your draft? That’s when nobody even called it “AI” anymore. It was just... what the workflow did. Faster closes, fewer errors, and nobody even noticed the “technology” part.

Mark Putnam

That’s the pivot people miss. At one institution I know, they sunk so much into this custom AI dashboard for admissions—showy stuff. But it was only after they gave up on the standalone glitz and plugged predictive analytics right into their admissions and retention systems that it clicked. Suddenly, the recruitment cycle ran smoother. It wasn’t about talking “AI strategy” in every meeting. Results spoke for themselves, and AI faded into the pipes just like reliable old electricity.

Olivia Carter

And it’s almost funny: when things truly work, nobody cares if it’s “powered by AI.” They just notice that things run without drama, and then everyone moves on. That’s when you know a technology’s really arrived.

Olivia Carter

But it wasn’t just about smoother workflow—governance got real, finally. In the early hype, everyone talked a big game about AI ethics and responsible use. But, honestly, most folks just copied some best practices off a competitor’s website and called it a day. Enforcement was, uh... minimal. That changed fast once the first lawsuits landed and regulators wanted receipts.

Mark Putnam

Yes. By 2035, what mattered was data lineage, audit trails, knowing precisely what your models did and when—not just waving around another “ethics statement.” Governance was now enforced, not aspirational. This also forced higher ed to face a critical insight: not every AI system is the same, not every risk is the same. Four domains emerged. Research infrastructure—high-stakes, patient investments with measurable payoff over time. Operational systems—deep integration that delivered visible value, especially to students and faculty. Staff productivity—the everyday stuff, AI in your email, in your spreadsheets, mostly about smarter orchestration, not magic. And, finally, teaching and learning—for students, AI shifted assignment design. The bar was raised; first-pass work was automated, so teaching shifted to focus on interpretation and synthesis.

Olivia Carter

It’s wild, Mark, but as you’re talking, I’m thinking of my alma mater. There was this data privacy audit, must have been 2029 or so? It uncovered—well, let’s say “creative” uses of data nobody really wanted to admit. That’s what finally triggered genuine accountability. Policies stopped being theoretical. Guardrails actually got built, and the tone changed overnight—it wasn’t just “AI is the future,” it was “how do we make sure this is safe, reliable, and fair for real people?”

Mark Putnam

That’s where progress really stuck. Institutions stopped trying to do everything, everywhere. They got specific about where AI belonged, what risks were tolerable, and where humans—real, trained professionals—must remain in the loop.

Mark Putnam

Charisma always feels crucial during the boom. The visionaries, the “AI-strategy evangelists”—their job was selling the sizzle. But that only gets you so far. When abundance gives way to scarcity and questions sharpen, only stewardship endures. What mattered after the AI bubble? Leaders who invested in credibility, discipline, and systems that could survive budget crunches. Not showy experiments, but the reliable stuff that carried you through stress. I used to tell my teams—lead for the era when budgets tighten and memories are long.

Olivia Carter

I keep coming back to that. Once everything’s noisy and flashy, it’s tempting to chase the new headline. But the institutions that really thrived post-bubble—whether universities or newsrooms or anyone else—were the ones who invested in their bones, not just in the glitz. And they made sure their teams could adapt and think, not just click a button.

Mark Putnam

Exactly. Legacy comes from building things that last beyond you, beyond the hype cycle. That’s what makes all the difference when the mood shifts and only the essentials count.

Olivia Carter

Alright, let’s leave it there for today. I feel like this episode could easily turn into a whole series—what lasts, what fades, and how we all keep moving forward. So thanks, Mark, and thanks to all of you listening. As always, we’ll be back soon with more dispatches from the frontlines of tech, leadership, and, uh, whatever else the next cycle throws our way!

Mark Putnam

Thanks, Liv. Take care, everyone—and remember, sometimes the quietest changes are the ones that stick. Until next time.