AI, Not AI: Riding the Hype Cycle – Episode 45

Agents at the Peak, Humans in the Loop: Navigating the AI Hype Cycle

Every week brings another “breakthrough” headline—agent modes, study modes, version bumps—and it’s getting harder to tell hype from progress. In this new Modem Futura episode, we take a candid, summer‑mode breather to map where AI really sits on the Gartner Hype Cycle, what open‑weight releases mean for builders, and how to keep the human voice intact when co‑authoring with machines. (Yes, we tried not to talk about AI…and failed—because it’s interwoven into every aspect of human activity now.)

What open weights really unlock

Setting aside the current drama around GPT-5, recent open‑weight releases under permissive licenses are a quiet game‑changer. OpenAI has released a pair of open‑weight models (120B & 20B) under Apache‑2.0 license that you can download from Huggingface. Translation: you can download models, run them locally, and adapt them for your own needs—no cloud required (except o download). With capable personal computers (think Apple’s M‑series) or home-built rigs GneAI LLMs can be run locally on device, and as hardware capacity increases and the sophistication of the models improves, the barrier to entry keeps dropping. The reason this matters is that it enables “garage‑scale” innovation—students, labs, startups, and curious tinkerers can now build for their own unique (or weird), local needs rather than waiting for a platform update.

Writing with AI—and protecting the voice

We also dig into human‑AI co‑authoring. Andrew shares a writer’s perspective—AI can draft moving, polished prose, but a subtle sameness creeps in. The fix isn’t anti‑AI; it’s pro‑craft: re‑introduce your “tells,” rhythm, and variance so readers feel a human mind at work. Think editorial sculpture—chipping away until the voice has texture and life. When even an AI editor flags your draft as “too consistent,” it’s a nudge to put the messiness back in. This is what happens when the pendulum swings too far to one side (perfect AI generated prose) the reader craves authenticity and “style” to which we need to introduce our human-touch back into the machine.

So…where are we on the Hype Cycle?
Whether you’re looking to learn how to interpret this powerful model (tool) or just get some new band name ideas, we explain the curve (innovation trigger → peak of inflated expectations → trough of disillusionment → slope of enlightenment → plateau of productivity) and why agentic AI feels perched at the peak, while day‑to‑day generative AIis edging into the trough—not because it’s useless, but because the shine (over hyped exaggerated claims of impact) wears off and the real work begins (just look at the backlash from GPT-5). Layer in the diffusion‑of‑innovation model and you’ll see different communities (VCs, educators, enterprises) living on different parts of the curve at the same time.

Image source: pasqal.com

Image source: Gartner

Beyond screens: ambient intelligence

We explore the exciting space of spatial/ambient computing and sensing (I even got to briefly mention LANs, WANs, and PANs)—environments saturated with signals (Wi-Fi, Bluetooth, cellular, NFC) that AIs can interpret in ways we can’t. It raises the question of what happens when machines can interpret the data‑saturated world beyond our comprehension and act within it? That’s where “AI‑not‑AI” lives: less chatbot magic, more embedded intelligence shaping everyday environments. That’s both exciting and unsettling: it demands new conversations about design, privacy, agency, and the futures we actually want to build.

If it resonates, help broaden the conversation: subscribe, share with a colleague, and tell us where you place AI on the Hype Cycle—and where you’re craving more human messiness. As we joked in‑studio, Modem Futura is “on the slope of enlightenment—accepting social investment via ratings and reviews

🎧 Listen on Apple Podcasts: https://apple.co/47mypmb

📺 Watch us on YouTube: https://www.youtube.com/@ModemFutura

If you’d like to dive deeper, jump into the link and listen to the podcast or watch the YouTube video. Join us as we explore the forces shaping our collective future and the urgent need to keep human values at the heart of innovation.

Subscribe and Connect!

Subscribe to Modem Futura on a favorite podcast platform, follow on LinkedIn, and join the conversation by sharing thoughts and questions. The medium may still be the massage, but everyone has a chance to shape how it kneads modern culture—and to decide what kind of global village we ultimately build.

🎧 Apple Podcast: https://apple.co/47mypmb

🎧 Spotify:  https://open.spotify.com/episode/4ReAdtrV7o8WfxeZ0vaKH9?si=dGnDFD03QiW9A3UiUStEEg

📺 YouTube: https://youtu.be/cfHqBJKnGZo

🌐 Website: https://www.modemfutura.com/

Summer School with AI: Rethinking Learning in the Age of GPT – Episode 41

Summer School with AI: Why “Back to Basics” Isn’t Enough

Why this episode matters? If you’re charting strategy for schools, workforce development or lifelong learning, this discussion offers a candid roadmap—and a few provocative questions—for navigating the next decade of educational transformation.

This month on Modem Futura I welcomed Rachna Mathur, Ed.D. —engineer, artist, lifelong-learner and Senior STEM Strategist at ASU Preparatory Academy—to a scorching‑hot Arizona studio for a free‑flowing “summer session” on the future of learning in the age of generative AI.

Our conversation touches on many aspects of learning and AI, but laser in on the implications of living the “digital world” for learning, partially inspired from the headline that shocked many educators: Sweden’s decision to pull back from screens and re‑embrace handwriting and printed books after seeing declines in comprehension and critical‑thinking benchmarks. We explore the move as an important—but incomplete—signal. We arguee that the real challenge is finding a sustainable balance between analog depth and digital acceleration, not retreating wholesale from technology, and not leaning into a pure technological solution just for technologies sake.

The theme of moderation threads the entire episode. We swap Montessori childhood stories—self‑directed, community‑anchored, and surprisingly common among tech leaders—before examining how that philosophy might translate to AI‑rich classrooms where personalization risks isolation if community norms aren’t protected.

We then fast‑forward 50 years to imagine two stark futures: a post‑scarcity Star‑Trek‑style society of flourishing creativity, or the WALL‑E “hover‑chair” dystopia where humans outsource thinking, writing and even curiosity to autonomous agents. In both scenarios, today’s policy and design choices in K‑12 systems carve the path. Should we double‑down on foundational literacies—or teach students how to audit machine output for bias, hallucination and relevance?

We highlight the rising cognitive load on teachers, who are expected to master every “shiny new doodad” while still wearing a dozen other hats. We discuss realistic guardrails: cell‑phone moderation policies; AI readers that empower dyslexic learners; and iterative, living guidelines that evolve alongside the tech itself rather than one‑and‑done declarations.

Finally, we confront the looming content‑collapse problem (the recursive nightmare that may be building right in front of us): models now train on data increasingly generated by other models, a self‑referential “snake eating its own tail” that threatens originality and human perspective. Our shared conclusion? Educators, parents and technologists must collaborate on a middle path that preserves human agency, cultivates critical judgment, and leverages AI as an amplifier—not a crutch.

🎧 Listen on Apple Podcasts: https://apple.co/4o6LwOc

📺 Watch us on YouTube: https://www.youtube.com/@ModemFutura

If you’d like to dive deeper, jump into the link and listen to the podcast or watch the YouTube video. Join us as we explore the forces shaping our collective future and the urgent need to keep human values at the heart of innovation.

Subscribe and Connect!

Subscribe to Modem Futura on a favorite podcast platform, follow on LinkedIn, and join the conversation by sharing thoughts and questions. The medium may still be the massage, but everyone has a chance to shape how it kneads modern culture—and to decide what kind of global village we ultimately build.

🎧 Apple Podcast: https://apple.co/4o6LwOc

🎧 Spotify: https://open.spotify.com/episode/1E9LMkkOTYvTJVwlb6Ey0F?si=ADMbNGEWSUW-Jdx-h0oBsA

📺 YouTube: https://youtu.be/cvxHCJxahlg

🌐 Website: https://www.modemfutura.com/

Summer Movies, liquid media, and alien AI languages – Episode 40

From Popcorn to Paperclips: What Summer Blockbusters Teach Us About an AI‑Shaped Future

With Phoenix broiling at 120 °F, we opened the studio door to a blast‑furnace breeze and a full house of ideas. Episode 40 of Modem Futura is nominally a “summer movies” chat—but the conversation quickly melts into a much richer alloy of cinema, ethics, pedagogy and speculative futures.

A spoiler‑free Superman (2025) debrief kicks things off. Sean relishes Dolby Atmos thunder and crowd‑pleasing cameos while Andrew savors the rare joy of a superhero film that is simply “incredibly fun” without the need to be anything but entertaining.

That sets the stage for the surprise gem of the season: M3GAN 2. (Or more specifically, Andrew’s revelation of how much he enjoyed it). Far from a Chucky retread, the sequel pivots into full‑blown techno‑thriller territory—surfacing neural‑chip debates, AI value‑alignment nightmares and invokes the infamous “paperclip maximizer” thought experiment. The hosts cheekily ask whether investing in AI is humanity’s salvation…or the catalyst for its conversion into stationery.

Movies as stealth pedagogy loom large. Andrew describes his film class where popcorn favorites become Trojan horses for serious reflection; students re‑watch titles with friends and family, eager to dissect ethics and innovation themes they can’t un‑see. Sean and Andrew discuss how the formula works because it lowers the barrier to entry while secretly building critical‑thinking muscle.

From here, just as the movies themselves act as Trojan Horses we get into some deeper ideas:

• Story archetypes rebooted. Are the classic five conflicts (character vs. self, society, nature, etc.) universal, or will alien machine intelligences invent a sixth form of narrative that we literally cannot grasp?

• Liquid media & the dead‑internet theory. When every asset can be remixed on demand, text becomes speech becomes video—and bots may already outnumber humans online. How do we preserve authentic signal in an ocean of generative noise?

• Chaos theory for a networked planet. Eight billion hyper‑connected humans + foundation‑model AI = a complex system hurtling toward new tipping points. Can we always innovate out of disruption—or does that curve eventually outpace us?

The episode closes with a cheeky pitch for Hollywood: “Clippy: Revenge of the Paperclip Maximizer.” Microsoft’s once‑loathsome office assistant becomes the perfect foil for an alignment‑gone‑wrong blockbuster—and a reminder that even silly artifacts can spark serious futures thinking.

Why it matters: Whether you’re an educator looking for sticky teaching tools, a technologist wrestling with alignment, or a storyteller hunting the next frontier, this discussion shows how pop culture can illuminate the biggest questions about being human in an AI age.

🎧 Listen on Apple Podcasts: https://apple.co/3GD73NA

📺 Watch us on YouTube: https://www.youtube.com/@ModemFutura

If you’d like to dive deeper, jump into the link and listen to the podcast or watch the YouTube video. Join us as we explore the forces shaping our collective future and the urgent need to keep human values at the heart of innovation.

Subscribe and Connect!

Subscribe to Modem Futura on a favorite podcast platform, follow on LinkedIn, and join the conversation by sharing thoughts and questions. The medium may still be the massage, but everyone has a chance to shape how it kneads modern culture—and to decide what kind of global village we ultimately build.

🎧 Apple Podcast: https://apple.co/3GD73NA

🎧 Spotify: https://open.spotify.com/episode/2BeXNdntLQlPwsUlKYnO7h?si=Xw7nG8HcQpyydGkb2TCeUQ

📺 YouTube: https://youtu.be/3kTVC4LHeYM

🌐 Website: https://www.modemfutura.com/

Futures of Learning: AI in Education with Punya Mishra – Episode 33

Friction Required: How will a world transformed by emerging technologies like AI reshape the world? Sean Leahy,Andrew Maynard and special guest Punya Mishra cut through the hype to reveal the creative tension, hidden risks, and big-picture futures for AI-powered, human-centered education. How can the power of AI be harnessed without losing the soul of learning?

Friction Required: Re-imagining Learning in an AI World

Generative AI burst onto campuses promising personalized tutoring, instant lesson plans, and anytime feedback. Yet beneath the buzz lies a more provocative question: What, exactly, makes education worth the effort once answers are a prompt away? In this week’s Modem Futura, hosts Sean Leahy and Andrew Maynard sit down with educator-innovatorDr. Punya Mishra to look past the shiny tools and into the messy, human heart of learning.

🎧 Listen on Apple Podcasts: https://apple.co/3ZDH8vg

📺 Watch us on YouTube: https://www.youtube.com/@ModemFutura

Over an energetic hour they explore why purposeful “friction”—the struggle, inquiry, and face-to-face negotiation of meaning—is still essential. Punya and Sean draw on John Dewey’s four impulses—Inquiry, Communication, Construction, Expression—as a compass for designing AI-infused classrooms that amplify (rather than automate) these deep-learning moments. The trio swap stories of chatbots that spark creativity, debate whether banning tools curbs cheating or curiosity, and ask whether transparency beats top-down rules when it comes to academic integrity.

But the conversation zooms further out. What happens when large language models become persuasive co-teachers? Could Universal Basic Income turn learning into a lifelong pursuit instead of a credentialing race? And might universities act as society’s “flywheel”—a deliberate drag that buys time to think before technology rewrites the rules? The answers aren’t neat, yet they underscore a shared conviction: the future of education must be AI-powered and human-centered.

Key Takeaways

• Friction is a feature, not a bug. Struggle fosters agency, resilience, and creativity—qualities that instant answers risk eroding.

• Design for Dewey’s impulses. Use AI to scaffold inquiry, amplify student expression, and make thinking visible, not to short-circuit it.

• Radical transparency > blanket bans. Open dialog about capabilities, limitations, and ethics beats whack-a-mole policies.

• Cheating vs. caring. Focus on cultivating authentic motivation; surveillance tech alone can’t fix a trust gap.

• Universities as sandboxes and speed-bumps. Higher ed can prototype responsible uses and slow premature adoption that harms society.

Whether you’re an instructor drafting next semester’s syllabus, a student exploring new creative tools, or a policymaker worried about the automation of learning, this episode offers frameworks—and questions—to keep humans at the center of the AI revolution.

🎧 Ready for the full conversation? Click below to listen or watch, then let us know how you’re embracing (or resisting) AI in your own learning spaces. And if the discussion sparks ideas, consider sharing this newsletter with a colleague—friction loves company!

If you’d like to dive deeper, jump into the link and listen to the podcast or watch the YouTube video. Join us as we explore the forces shaping our collective future and the urgent need to keep human values at the heart of innovation.

Subscribe and Connect!

Subscribe to Modem Futura on a favorite podcast platform, follow on LinkedIn, and join the conversation by sharing thoughts and questions. The medium may still be the massage, but everyone has a chance to shape how it kneads modern culture—and to decide what kind of global village we ultimately build.

🎧 Podcast: https://apple.co/3ZDH8vg

📺 YouTube: https://www.youtube.com/@ModemFutura