DL 443
A Seat at the Table
Published: August 12, 2026 β’ π§ Newsletter
Hi all, welcome back to Digitally Literate.
This week's issue is about who gets to decide what AI becomes, and how rarely that decision includes the people who'll actually live with it. Three groups tried to answer that question this week: teenagers, billionaires, and everybody standing next to a half-built server farm. Only one of them produced anything you could call a rule.
It's an old argument, dressed in new infrastructure. A century ago it was railroads. Engines of progress to the people who owned them, and someone else's rate schedule to everyone downstream of the tracks. The technology changes. The question of who gets a say in it doesn't.
As always, your support is valued. Reach out anytime at hello@wiobyrne.com, and subscribe if you haven't already.
The Kids Wrote a Bill
Ninety-eight high schoolers, one from every state, spent three days in July inside a precise replica of the U.S. Senate chamber at the Edward M. Kennedy Institute in Boston. They weren't touring it. They were using it, drafting an actual bill on how schools should handle AI, then voting it through, 82-16.
The result, the Students First Act, isn't complicated. Teach AI literacy as soon as a student gets a school device. Ban AI on graded tests. Let teachers require a student to defend their work out loud if AI use is suspected. It took the only people in this fight who've never gotten a vote, the students, three days to produce what a year of adult task forces hasn't.
They know exactly what they're responding to. "We missed the mark with social media and phones," said Hunter Wurzel, a senior from Ohio, naming the last technology adults failed to govern before handing it to kids anyway. Bena Mejia Romero, from Connecticut, put it more plainly: "This bill is about who gets a seat at the table. Today, students have that seat." Jeffrey Riley, who used to run Massachusetts's schools and now leads Day of AI, was blunter still: the adults couldn't agree, "so now it falls to the kids."
This isn't law. It's a model text, and AASA is mailing it to 10,000 superintendents who are writing their own AI policies right now, mostly from scratch. Whether it goes anywhere depends on whether any of them pick it up.
When the group with no legal standing outperforms the group with all of it, that's not a feel-good story. That's a diagnosis.
Billionaires Wrote Manifestos
On August 10, Mark Zuckerberg published a 6,500-word essay called "The Future is for Everyone." In it, he details what he calls "a philosophy for how we can best use superintelligence to ensure it improves all of our lives, work, communities, freedom and safety." In it, he asks "who will have access to superintelligence and what will we direct it towards. Will it be centralized and restricted to a few institutions, or will it be a tool that empowers everyone?" His answer is free access for billions, and for everyone else, "a dynamic auction mechanism that will guarantee that everyone gets the lowest price possible." In 2019, he published a different essay, "The Future is Private," while running an ad business built on the opposite.
It's worth remembering who's talking. Meta owns Facebook and Instagram. For a huge share of the world, those two apps are the internet. Meta also offers Llama, the most widely used open-weight AI model family. And Meta has a track record. What the company says it's doing and what it's actually doing tend to diverge, sometimes completely. Facebook and Instagram have already started filling feeds with AI-generated content to hold attention, whether or not anyone asked for it. Put those two facts together and the essay reads differently. It's not really a philosophy of superintelligence. It's a preview of coming attractions. For a large fraction of the planet, "AI" isn't going to be a choice they make. It's going to be whatever Meta decides to put in the feed.
This week I also came across Anthropic CEO Dario Amodei's outline for the future and it's given me some pause. Back in January, Amodei warned in "The Adolescence of Technology" that AI could create a permanent "underclass," people who lose both income and political leverage as AI substitutes for labor. He's since reaffirmed his prediction that half of entry-level white-collar jobs could vanish within one to five years. This post reads like a warning, but it's also something you'd expect to hear from someone selling AI. Urgency does wonders for the person selling it. "Adapt or get left behind" has been the pitch of every technology cycle back to electricity.
Fifteen months earlier, in October of 2024, Amodei also wrote almost the opposite post in "Machines of Loving Grace. In it he imagines AI as a "country of geniuses in a datacenter," compressing a century of medicine into five to ten years. Most cancer gone, most mental illness treatable, human lifespan doubled. Amodei opens that essay promising to avoid "grandiosity" and "sci-fi baggage," then spends 15,000 words describing something close to both.
The Public Writes Refusals
For me, one of the top stories about the AI boom focuses on data centers. I recommend this piece from Jasmine Sun to get more insight.
No one has been asked if they want datacenters in their neighborhoods, but people are showing up and making their voice heard. More than 200 local and state bans or moratoriums on new datacenters took effect in the last 30 days alone, bringing the active total nationwide past 550, up from around 300 in late June. New York's governor signed a one-year pause on permits for facilities over 50 megawatts in July. Texas's governor ordered the grid operator to audit nearly 500 gigawatts of pending datacenter connection requests before approving any more. Bernie Sanders and Alexandria Ocasio-Cortez introduced a bill to freeze construction nationally.
The numbers behind it help tell the story. A Heatmap News poll found opposition to a nearby datacenter jumped from about 4 in 10 people last August to 7 in 10 by this May. Gallup found that 71% oppose a datacenter next door, but only 53% oppose a nuclear plant next door. People are more afraid of the AI server farm than the reactor that powers it.
Keep in mind, none of this is simple. Some of these bans are already getting sued out of existence, and plenty of towns are counting on datacenter tax money to balance their budgets. We see examples of people standing up and speaking out, and then the local government goes against their wishes and approves the project anyway. For example, in Saline Township, Michigan, they voted overwhelmingly (4β1) to reject rezoning 575 acres of prime farmland for Related Digitalβs $16 billion, 21-million-square-foot OpenAI/Oracle data center. Local officials and residents cited environmental concerns, noise, traffic, and conflicts with their master plan.
However, Related Digital immediately sued, alleging "exclusionary zoning." Facing costly litigation and limited legal leverage, the township quickly settled, allowing groundbreaking to proceed in exchange for $14 million in community benefits. The rapid approval sparked public outrage and a recall effort against board members, underscoring how corporate leverage and state land-use laws can override small-town local zoning decisions.
The Understory
In the late nineteenth century, the railroads were sold as engines of progress, stitching a vast country together and opening the West to commerce. But for the farmers living along those tracks, progress came with a price. Railroad companies seized land, dictated what it cost to move crops to market, and wielded enormous influence over state governments. A farmer might pay more to ship grain fifty miles to the nearest depot, where the railroad had no competition, than a rival paid to ship it five hundred miles to Chicago, where it did. The rate wasn't about distance. It was about who had leverage.
Farmers responded by organizing. The Grange, and later the Populist movement, turned local frustration into demands for public control over private infrastructure. Illinois and several other states passed "Granger Laws" regulating railroad and grain-elevator rates. In 1877 the Supreme Court upheld Illinois's right to do it, ruling that a business "clothed with a public interest" could be regulated by the state even without the owner's consent. The federal government didn't act until a decade later, with the Interstate Commerce Act of 1887, and only after a later ruling showed states couldn't reach traffic that crossed their borders on their own. States moved first. Washington moved last, and only when it had to.
That's the shape of this week, too. A handful of people are writing the philosophy of who should control the country's most powerful new infrastructure, the same way the railroad barons wrote about progress and civilization while setting their own rates. The people living next to that infrastructure are organizing state by state, the way the Grange did, well ahead of any federal answer. And the students who wrote their own AI policy did what farmers did in the 1870s. The people closest to the problem stopped waiting for institutions to solve it and wrote the rules themselves.
The 1887 lesson was that infrastructure powerful enough to shape public life can't be left entirely to the people who own it. The 2026 version of that lesson hasn't been learned yet. Right now, it's optional
See you next Wednesday. As always, my email is hello@wiobyrne.com.
Digitally Literate is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber.
π Navigation
Previous: DL 442 β’ Archive: π§ Newsletter
πΈοΈ Connected Concepts:
- AI Governance β Who gets to set the rules for a powerful new technology, and whether that authority sits with the people building it, the people regulating it, or the people living with it.
- AI Populism β Treating AI as an elite project imposed without consent, rather than a technology to be judged on its merits.
- AI in Education β How schools, teachers, and students are being asked to adopt AI faster than any institution has produced guidance for it.
- Tech Manifestos β The genre of essay in which a company's leadership describes the future its own product will supposedly bring about.
- Infrastructure Politics β The fight over who bears the physical, environmental, and financial cost of building the infrastructure a new technology requires.