The Waning Days of Techno-Futurism
Reorienting our relationship with AI through the lens of technoskepticism.
It has been a long time since I’ve written in a sweeping way about the state of AI sentiment. In the early days of this Substack, I leaned heavily on a tripartite schema: AI optimism, pragmatism, pessimism. It was a way to make sense of how people were positioning themselves in the wake of ChatGPT’s arrival. Useful scaffolding at the time. But I have learned a great deal since then, and the schema no longer holds.
The truth is that this work is far too complicated for clean categories. Most of us occupy all three positions over the course of a single day, sometimes multiple positions simultaneously. You can marvel at what a model produces in one breath and feel a pit in your stomach about what it means in the next. You can be pragmatic about integration at 9 a.m. and deeply pessimistic by lunch. These are not contradictions. They are the lived experience of people trying to think honestly about a technology that is reshaping the ground beneath their feet in real time.
So I want to take a step back and try, once more, to characterize the landscape. Not with tidy boxes, but with the texture of what has actually happened since the early fever broke.
The Manifesto That Marked the Peak
In October 2023, Marc Andreessen published what he called “The Techno-Optimist Manifesto,” a 5,200-word declaration of faith in technology as the singular engine of human progress. He used the phrase “We believe” over a hundred times. He called AI a “philosopher’s stone.” He listed 56 intellectual “patron saints.” He dismissed trust and safety, sustainability, and tech ethics as enemies of progress. He compared universal basic income to farming humans like zoo animals.
The manifesto was grandiose, quasi-religious in structure, and utterly certain. Critics called it everything from a Nicene creed for the cult of progress to old-school reactionary elitism dressed up in Silicon Valley vernacular. But its real significance, I think, is less about its content than its timing. It marked a kind of high-water point, the moment when techno-futurism felt bold enough to declare itself openly, without hedging, without apology.
That confidence has since eroded considerably.
The Sentiment Shift
If 2023 was the year of AI’s awakening and 2024 was the year of frantic adoption, then 2025 was the year the backlash crystallized into something real. And 2026 is shaping up to be the year when that backlash becomes a political force.
The data tells a stark story. Pew Research Center's June 2025 survey of over 5,000 U.S. adults found that 50% are more concerned than excited about AI's growing role in daily life, up from 37% in 2021. Only 10% say they are more excited than concerned. A separate Pew study found that while 47% of AI experts are more excited than concerned, that number drops to just 11% among the general public. Fifty-seven percent of Americans rate the societal risks of AI as high; only 25% say the same about its benefits. Internationally, across 25 countries surveyed, in no single country does the largest share of adults say they are more excited than concerned. The concern is bipartisan, cross-generational, and global.
But the numbers only tell part of it. What has actually changed is the texture of daily life since 2023.
AI text has engulfed the internet. Over a thousand news sites are now run almost entirely by bots. Google’s own search results are inundated with pages that, as the company itself acknowledged, feel like they were built for algorithms rather than people. The “dead internet theory,” once dismissed as fringe conspiracy, has entered mainstream discourse as something closer to observable reality. AI-generated text floods social media, poisons search results, and degrades the information commons in ways we are only beginning to reckon with.
Data center protests have erupted across the country. From Virginia to Arizona to Michigan, communities are organizing against the physical infrastructure of the AI boom. In the second quarter of 2025 alone, opposition blocked or delayed an estimated $98 billion in data center projects. Over 180 activist groups now operate across 17 states. A town in Wisconsin tried to oust its mayor over a data center approval. The Washington Post reports that data center resistance is fast emerging as a potent electoral issue, bipartisan, visceral, and local. Residents are not debating abstract AI ethics. They are fighting about water, electricity bills, noise, and the industrialization of their neighborhoods.
Politicians routinely deploy AI-generated fakes as political instruments. During the first months of his second term, President Trump posted AI-generated images of himself as the Pope, as a Jedi, and flying a fighter jet labeled “King Trump.” AI-generated deepfakes have been used in elections across Ecuador, Germany, Romania, and the Netherlands. In Indonesia, AI was used to reanimate the dead dictator Suharto to endorse a political party. What was once a speculative concern about deepfakes has become a mundane feature of the political landscape, not a crisis event but a constant background hum.
Schools, now three years into this, find themselves in genuine crisis. A 2025 survey found that 88% of students reported using generative AI tools for assessments, up from 53% the prior year. Faculty describe feeling less like instructors and more like detectives. AI detection tools are unreliable, producing false positives that damage trust and false negatives that render the tools useless for high-stakes decisions. Institutions are scrambling toward oral exams, in-class writing, and proctoring pilots, essentially retreating to pre-digital assessment formats, while still lacking coherent AI policies, meaningful professional development, or any shared framework for what “learning” even means when a machine can produce a passable essay in seconds. The curriculum gap is real: content-area standards still do not explicitly address generative AI, and most educators are working without disciplinary guidance.
None of this was supposed to happen so fast. Or rather, the technology was supposed to arrive fast, but the consequences were supposed to wait politely in line.
The DSAIL Cohort and What Technoskepticism Has to Offer
This brings me to work that is close to my heart.
Everything I described above, the degraded information landscape, the political weaponization of synthetic media, the crisis in schools, can feel paralyzing when encountered alone. One of the things I have come to believe is that the only way through an ongoing crisis like this is together, in sustained conversation, with people who are willing to sit in the difficulty rather than rush toward easy answers.
That is what the DSAIL cohort is. Now four meetings strong, this group has become a model for what it looks like to work through struggle rather than around it. The members show up consistently. They bring their disciplines, their classrooms, their doubts, and their questions. They are not chasing the next tool or the next prompt hack. They are doing the slower, harder work of figuring out what generative AI means for teaching, for learning, for the integrity of the knowledge we pass on to students. In a landscape saturated with hot takes and hype cycles, that kind of sustained, human, collaborative work is exactly what no algorithm can replicate or replace.
And the cohort is already producing scholarship that matters. Samantha Serrano, a veteran geography and civics teacher from the northwest suburbs of Chicago and a member of the DSAIL cohort, recently published a piece in the Journal of Geography that I think deserves wide attention. Grounded in the technoskepticism articulated by Krutka, Heath, and Mason (2020), Serrano builds a full typology of generative AI errors: Fabrication, Semantic, Flattening, Media Representation, and Contextual Offense. She maps each domain against the epistemic lenses that undergird the social studies (Space, Time, Human Experience, and Societal Systems), giving educators a critical vocabulary for naming what goes wrong when probabilistic language models are used to construct disciplinary knowledge.
Technoskepticism, as Krutka et al. originally framed it, does not reject technology. It interrogates it. It asks what a society gives up for the benefits of a technology. Who is harmed and who benefits. What the technology needs from us. What unintended changes it causes. Why it becomes difficult to imagine our world without it. These are not anti-technology questions. They are the questions that social studies education was built to ask.
What Serrano does is take those questions and apply them directly to the structural limitations of large language models: their horizontal logic, their inability to trace information to a verifiable source, their tendency to flatten chronology and homogenize culture. Her examples are grounded in the kinds of errors that classroom teachers will actually encounter. An AI-generated map that places a 16th-century city in the Bronze Age. An LLM that produces a stereotyped monologue when asked to simulate an Ainu elder. Outputs that rank the UN Sustainable Development Goals with normative authority as though they were empirical facts.
This is not abstract theorizing. It is disciplinary-specific critical work, the kind the field desperately needs as GenAI becomes embedded in lesson planning, instructional materials, and student research workflows. And it is exactly the kind of work that emerges when educators have a community to think within, when they are not navigating this crisis alone.
Closing
Andreessen’s manifesto declared that we are “the masters of technology, not mastered by technology.” That framing, dominion, control, mastery, feels increasingly disconnected from the reality that most people are actually living. People are not mastering AI. They are navigating it, mitigating it, resisting it, accommodating it, and trying to figure out what it means for their work, their students, their communities, and their sense of what is true.
The waning of techno-futurism is not the arrival of techno-pessimism. It is something more interesting and more difficult: the slow, uneven emergence of a critical posture, one that refuses both uncritical adoption and reflexive rejection in favor of harder questions about what this technology actually does, who it serves, and what it costs.
Serrano’s typology, Krutka’s technoskepticism, the DSAIL cohort’s steady work: these are not headline-grabbing developments. But they represent, I think, the intellectual infrastructure we will need for whatever comes next.
The fever is breaking. Now the real work begins.
Check out some of our favorite Substacks:
Mike Kentz’s AI EduPathways: Insights from one of our most insightful, creative, and eloquent AI educators in the business!!!
Terry Underwood’s Learning to Read, Reading to Learn: The most penetrating investigation of the intersections between compositional theory, literacy studies, and AI on the internet!!!
Suzi’s When Life Gives You AI: A cutting-edge exploration of the intersection among computer science, neuroscience, and philosophy
Alejandro Piad Morffis’s The Computerist Journal: Unmatched investigations into coding, machine learning, computational theory, and practical AI applications
Michael Woudenberg’s Polymathic Being: Polymathic wisdom brought to you every Sunday morning with your first cup of coffee
Rob Nelson’s AI Log: Incredibly deep and insightful essay about AI’s impact on higher ed, society, and culture.
Michael Spencer’s AI Supremacy: The most comprehensive and current analysis of AI news and trends, featuring numerous intriguing guest posts
Daniel Bashir’s The Gradient Podcast: The top interviews with leading AI experts, researchers, developers, and linguists.
Daniel Nest’s Why Try AI?: The most amazing updates on AI tools and techniques
Jason Gulya’s The AI Edventure: An important exploration of cutting-edge innovations in AI-responsive curriculum and pedagogy
Excellent framing, Nick. Thank you!
Thanks for another great essay! FWIW, I introduce DSAIL in every talk I give in higher education. I find that it neatly walks the line between resistance and embrace.