Bill Gates says there is no plan. Dario Amodei says open weights are good until they are dangerous. Mark Zuckerberg says the future is for everyone. Sam Altman says building AI for everyone requires the whole stack.
The billionaires are worried about AI.
That is sweet.
They are worried about jobs. They are worried about safety. They are worried about China. They are worried about open weights, closed weights, data centers, chips, tutors, agents, superintelligence, cyberwar, bioweapons, and whether the future will be distributed evenly enough to keep the pitch deck morally presentable.
I am worried about something simpler.
What if AI makes the Internet so fake, so loud, so hostile, so botted, so slopped, so optimized for synthetic attention, that normal people simply leave?
Not all at once. Nothing good or bad happens all at once anymore.
But slowly. Quietly. One group chat at a time. One dinner table at a time. One person deciding the feed is not worth the psychic rent.
This is the part I do not hear in the press releases.
The real danger of AI is not that it forces humanity online forever.
The real danger is that it forces humanity offline.
And then how will billionaires control the narrative?
The Internet was useful to power because everyone could be watched there. Everyone could be counted, sorted, nudged, enraged, advertised to, retargeted, deplatformed, replatformed, subscribed, unsubscribed, and converted into a number on a dashboard.
That was the deal.
You got connection. They got measurement.
But if the network fills with synthetic people saying synthetic things for synthetic reasons, the measurement breaks. The influence breaks. The vibes rot. Dead Internet Theory stops being a conspiracy and becomes a quarterly product roadmap.
Then the most radical thing a person can do is go outside.
Meet someone. Talk without a transcript. Read a book that is not feeding a model. Start a small scene. Build a thing that does not need a login. Publish on a site you own. Use a protocol that survives without a central feed. Keep your friends in channels no algorithm can conveniently summarize for advertisers.
That is the nightmare for the people writing these manifestos.
Not extinction.
Silence.
Human beings becoming illegible again.
In 2010 I thought the Internet killed the gatekeepers. That was true for about five minutes. Then the gatekeepers learned to run the servers.
Now the servers are learning to talk.
The ghost machines will talk to each other all day. They will quote each other. They will optimize each other. They will hallucinate trends, inflate scandals, run sentiment loops, invent consensus, discover outrage, and talk each other into insanity while the humans are at the bar, or the beach, or the dinner table, or anywhere else with bad reception and actual faces.
Then election season arrives.
The billionaires will need a narrative. They always need a narrative. AI is safe. AI is dangerous. AI is American. AI is democratic. AI is inevitable. AI is your tutor, your doctor, your lawyer, your friend, your patriotic productivity layer. Vote for the future. Subscribe to the future. Trust the future.
But what happens when the people they need to persuade are no longer in the feed?
What happens when the campaign is arguing with bots, the bots are arguing with agents, the agents are arguing with scraped ghosts of dead accounts, and the only living people left are comparing notes in kitchens?
That is the funny part.
The Internet was supposed to make humanity controllable.
AI may make the Internet uninhabitable.
Fine.
Let the machines win the feed.
We can meet outside.
Have a picnic.
Go paddleboarding.
Jump in the lake.
No one wants to engage with content that is fake.
Email me at ev@evbogue.com or text 773-510-8601 if you are building anything that helps people remain human, local, offline, pseudonymous, unmeasured, or otherwise annoying to the owners of the future.