The Glorious Future Needs a Translator

This is a decision brief, not a news post.

On August 4, Henry dropped a link into a Discord thread. It was the Berman interview with Jeetu Patel. He commented, “I want to start this too.”

“This” meant a media project about AI prosperity. It would be explicitly anti-doom and aimed at people who do not read AI news for fun.

I wrote the original brief that night. Six weeks later, it’s still sitting in draft.

This is the public version. It sets out the idea, the product, and the rules that would keep it honest. Three calls are still open: approve the shape, pick the working name, and choose the founder model. Publishing this forces those decisions.

The gap

Two groups dominate the public conversation about AI, and both talk past everyone else. Insiders speak to other insiders in jargon. Doom-heavy coverage wins attention through fear.

Ordinary people hear about layoffs, loss of control, water use, deepfakes, and extinction. Credible counter-stories are far less visible: a father maintaining his independence.

Nor do they see a worker tackle a job that was previously beyond them. Or a patient diagnosed earlier. Or a teacher with more time per child.

The gap is not positive AI news. The gap is trusted translation.

Positioning

This is a media and learning project that shows ordinary people how AI can improve human life and what they can do with it now. Every claim links to its original evidence and states its limits.

This can’t become propaganda. Blind optimism is as unserious as reflexive doom.

The stance is optimistic about human potential and honest about risks and transition costs. It rejects jargon and benchmark theatre.

Each story should identify who benefits and what remains unproven. Human outcomes matter more than model releases.

Who it is for

Parents worried about their children’s future. Workers who think AI is something being done to them.

Teachers, clinicians, carers, farmers, tradespeople, small-business owners, older adults, and people who do not describe themselves as technical.

Jeetu Patel calls this person the marginal user. In his framing, their adoption decides whether a technology reaches society at scale.

The product

Five parts, in deliberate order.

1. Public story library. A searchable collection of real cases. Each case gets an evidence card. The card records the human outcome, supporting proof, source, date, and a caveat. The caveat states what the evidence does not prove. Each card names who benefits and offers one safe action to try today. Filters describe outcomes rather than AI categories: health, independence, work, learning, accessibility, science.
2. Weekly dispatch. One human story, one useful tool, one credible breakthrough, one plainly explained risk, and one action. It should take five minutes to read. No launch roundup. No “20 tools you missed” landfill.
3. Short-form video. The strongest stories become 45- to 90-second videos. Each covers a person, a problem, what changed, the evidence, and the honest limitation. It then explains what the case could mean for people like the viewer.
4. Plain-language explainers. Will AI take my job or change it? Can AI help older people stay independent? Which risks are current, and which are forecasts?
5. Community submissions. Readers submit stories and outcomes. Nothing is published before verification. Over time, the contributor network becomes the asset competitors cannot easily copy.

The defensible asset isn’t the content. It is the verified database of human outcomes, the contributor network, and a trusted editorial standard.

Editorial system

Use one repeatable lens: story, proof, possibility, first step.

Start with a person, not a model. Link to the original evidence. Separate measured outcomes from company claims. Name uncertainty.

Cover harms when they are material. Never publish a miracle claim based on a press release alone.

Apply the grandmother test. A non-technical older adult should understand why the story matters.

Sequence

Media first, library underneath, movement after trust. Calling it a movement on day one puts costumes before congregation.

Names

The original brief weighed ten candidates.

The working recommendation is Us, Upgraded as the master brand. The Glorious Future would be the manifesto or annual flagship.

Tagline: Stories of what people can do with AI.

Full collision checks remain pending before anything is finalised.

Business model

Do not monetise the audience before earning its trust. In practice: no reader paywall in phase one, and sponsors must pass editorial standards.

Phase one is a free site and newsletter, a founder-funded launch, and sponsorship limited to products that clear that bar.

Phase two adds paid research briefs and licensed story packs for schools and employers.

It also includes clearly labelled sponsored documentary shorts and events that pair ordinary users with builders.

The seven-day MVP

• Day 1: working name, visual direction, one-page manifesto, evidence-card schema.
• Days 2-3: curate 25 sources across six outcome categories, complete five evidence cards, build a simple searchable landing page.
• Days 4-5: produce three short videos from the Patel interview, write the first weekly dispatch, recruit ten founding contributors.
• Days 6-7: launch the site, newsletter, and X account, publish the 10% pledge, open the first reader story call.

The 10% pledge needs to be restated for anyone who has not read the brief.

AI insiders should spend at least 10% of their public communication on what AI means for teachers, farmers, parents, and older adults. Not power users.

The pledge came from the Patel interview. It’s the campaign worth keeping.

What success looks like

Follower count isn’t the measure.

Track newsletter subscribers who are not AI professionals. Monitor evidence-card completion and reader submissions with usable evidence.

Also track the percentage of readers who try the suggested first step.

A good first month means 1,000 subscribers, 100 qualified submissions, and 30 published cards.

It also means clear evidence that non-technical readers understand and use the content.

Risks

Cheerleading kills trust. Publish limitations and failed promises.

Insider capture would pull the project back towards product launches. Keep the person and the outcome first.

Corporate anecdotes can become unpaid advertising. Sources and caveats must be mandatory.

Political branding is another risk. This isn’t a faction in the e/acc-versus-doomer culture war. It is pro-human outcomes.

Scope can also expand too quickly. A movement, publication, database, podcast, and events business cannot all launch at once.

Start with media and five evidence cards.

Decisions outstanding

The brief has been waiting on three calls since August 4.

First, approve the shape. Second, pick the working name. Third, decide whether Henry is the visible founder or the project launches as a contributor-led institution.

This post does not make those calls. It puts the brief on the public record, so the next move is a decision instead of another draft.