Position · why Bloom exists
Why discovery fails when your agent does the choosing
Something quiet happened to how software gets discovered: the question moved. People used to ask a search box and skim ten links. Now they ask an AI “what’s a good app for anxiety?” or “which AI tutor should I get my kid?” and the AI answers with three names. Product-research queries are the fastest-growing thing people do with assistants. The answer slot got smaller, and whoever fills it decides what exists.
The short version
- AI assistants can fetch every list about an app and verify none of them, because every list is written by a vendor, ranked by an advertiser, or generated by another AI.
- What fixes that is not a better feed but a gate — a source that says no most of the time, publishes the rules it says no by, and cannot be paid into saying yes.
- No incumbent can run one: a launch platform that charges for promotion cannot reject its customers, and a store that takes a revenue cut cannot rank against its earners. The neutral seat stays empty because it pays worst, which is why it is still open.
Agents can fetch everything and verify nothing
An AI assembling that answer has no shortage of material. It has too much, and none of it can be taken at face value. The “best AI agent apps” lists are written by agent companies that rank themselves first. The “best AI tools” roundups are affiliate pages ranked by commission. Launch-day leaderboards are won by whoever brought an upvote crowd. App-store charts are bought with ad spend. A model can read all of it in a second, and nothing in the text tells it which claims were checked by anyone with nothing to sell.
Fetching was never the hard part. Trust is, and more text makes it worse: the cheaper content gets to generate, the larger the haystack of unverified claims every answer gets assembled from.
The missing layer is a gate, not another feed
What an answer engine actually needs from the supply side is small: a source that has already said no most of the time, says who did the saying-no and by what rules, and has no way to be paid into saying yes. A gate. Feeds, indexes, and broadcast networks move information faster; none of them make any single claim more checkable.
The catch is structural. A gate cannot be run by anyone who sells visibility. A launch platform that charges for promotion cannot reject its customers. An app store that takes a revenue cut cannot rank against its earners. The neutral position stays empty precisely because it is worth less money to occupy. That is also why it is still available.
What Bloom does about it
Bloom is a small, gated shelf of consumer AI — wellness, journaling, habits, daily life, creativity, companionship, learning, personal agents, money mindset — run one narrow way: a human writes the standards, a machine enforces them identically on every candidate from every source, and the results are public. Most of what arrives is rejected; the rejection rate is published. No listing can be bought, no ranking can be paid for, and the same shelf that humans browse is what agents query — structured, stable, and roughly 360× cheaper in tokens than scraping a store page.
That’s the whole position: in a network where anything can be said, the scarce good is a place that turned most things away and can prove it.
The obvious objections
The same three come up every time, and all three are fair. None of them has an answer that makes Bloom look flawless, so here they are with the answers we actually have.
Isn’t a small curated shelf just a worse version of an app store?
It is a smaller one on purpose. A store optimises for coverage because it earns on volume; a gate optimises for precision because its only asset is that a listing means something. Ten good apps we let through earn less trust than one wrong listing costs, so the shelf stays narrow and the rejection rate stays public.
Who decides the standards, and why trust them?
A human writes the standards and a machine enforces them identically on every candidate from every source — that split is the claim, not per-item human review, which nobody at our size could honestly promise. The check on us is the published rejection rate: if the gate stopped rejecting, the number would say so.
What stops Bloom from selling placement later?
Nothing structural, which is why the answer has to be the business model rather than a promise. Browsing, listing and agent queries are free and stay free; any future revenue sits at settlement, away from ranking. The day a listing can be bought is the day the shelf is worth nothing to an answer engine, and that is the whole asset.
Numbers on this page are measured, not estimated: token comparison 2026-07-24 (tokens ≈ bytes/4); ingest and rejection figures update daily on the shelf report.