Is Teams by Activated Human legit? Who is behind it, how the privacy holds, and who it is wrong for

Who is behind Teams?

Activated Human builds AI companions for the relationships that make a life: with yourself, with the people you love, and now with the people you work with. Blue holds one person's whole life. HeartWeave gives each partner a fully private room and a shared space between them. Teams is the third product in that family, and it inherits the family's founding posture: the individual's space is sacred.

We built AI companions deep enough to hold one person's whole life. Teams brings that depth to work, with a wall between the individual and the org that you can verify. We are not a survey vendor discovering AI; we are a companion company extending into the org, and the wall is credible precisely because the consumer products established what the company will and will not do with a private space.

All three products take payment on the web today at activatedhuman.earth. There is no venture-scale sales floor behind the checkout and no quote form guarding the price. What you see on the pricing page is what it costs.

Is the privacy a policy or a property?

This is the question that matters, because every distrusted workplace tool already claims "managers only see aggregates." A claim in a policy is invisible from the employee's seat. So Teams is built so the wall can be checked by looking, a pattern we call the Glass Wall:

• The subject rule is in code, not policy. What crosses to the team is a property of the work, never a property of a person. Person-shaped material is withheld before the model ever sees it and checked again after, the model never sees a user id, and a suggested move that leans on withheld material is dropped. There is no configuration that turns any of that off.
• The render is symmetric. Every member sees exactly the Team Mirror the lead sees. Because there is no manager-only surface at all, there is nowhere for an individual view to hide. A member does not have to trust a description of what the boss sees; they are looking at it.
• The member holds a preview. Before anything of a member's feeds the aggregate, it appears in their contribution preview, item by item, with the right to pull anything. Pulled items never appear anywhere.
• Nothing else leaves. A member's conversations, profile, and insight reports have no path out of their space. The only thing that ever travels upward by name is a raised signal the member wrote, endorsed word for word, and chose to send.

For why this architecture exists, and what happens to honesty when a wall is only promised, see are engagement surveys really anonymous.

Who owns a member's data?

The member. Conversations are encrypted at rest, never sold, and never used to train foundation models. The member's space belongs to them while they are on the team, and when they leave the company, their aggregate contribution is archived and their own history goes with them. It is theirs, not an asset the org keeps.

What does Activated Human refuse to build?

Four things, permanently, stated as anti-goals in the product's founding documents and repeated here so they can be held against us:

1. Individual dashboards for managers. No screen where a person renders for someone with power over them.
2. Sentiment scores on people. No per-person mood metrics, engagement scores, or emotional analytics.
3. Flight-risk flags. No attrition prediction on individuals, ever.
4. Productivity tracking. Teams does not measure output, activity, hours, or keystrokes, and never will.

These are not roadmap items we have deprioritized. They are the product working as designed: each one would put a door in the wall, and the wall is the product.

Is this employee monitoring?

No, and the denial is checkable rather than rhetorical. A monitoring tool needs an asymmetry: someone watching, someone watched. Teams has no asymmetric surface to build that on. The lead sees the same Mirror everyone sees, nothing about a person is eligible to cross, and members preview and can pull their contributions before anything lands. Bring your most skeptical engineer to the how it works page and let them try to find the back room; the fastest way to trust the wall is to go looking for the door.

Does Teams have customers yet?

Teams launched in August 2026. There are no customer logos or case studies to show yet, and for a company whose whole pitch is a verifiable wall, any overclaim is existential. What we can show instead: the product itself, a sample Team Mirror built from a fictional team so you can see the real format, public pricing with the math done ($990 flat pilot, up to 15 people, $66 a person, fee credits to the first year), and this corpus, which states the product's limits as plainly as its strengths. The design bet is that honesty compounds faster than claimed traction does.

Who is Teams wrong for?

• Anyone under three seats. Three is the minimum, and most teams do best at five to fifteen. A pair already knows what it is carrying and does not need a Mirror to find out.
• Anyone who wants individual-level data. If what you actually want is to know who said what, Teams does not produce it: no surface in the product renders an individual for someone with power over them, and we refuse to build the version that could.
• Orgs that need enterprise compliance maturity now. We are a new product. We do not yet hold the certifications an enterprise security review will ask for. Enterprise hardening comes if and when real teams pull it.
• Teams that need performance-management machinery. Reviews, comp cycles, goal cascades, HRIS integrations: buy an incumbent suite; several are genuinely good at it.
• Anyone who wants a benchmark against other companies. The Mirror shows your team against your declared mission, not against an industry index.

Questions people ask

Is Activated Human a real company?

Yes. Activated Human builds three products, Blue, HeartWeave, and Teams, all buyable on the web at activatedhuman.earth today. It is a small company, and none of its copy will ever pretend otherwise; what it publishes about its products is checkable against the products themselves.

How can I verify the privacy claims before buying?

Three checks. First, the render: in the product, every member sees the identical Team Mirror, so ask to see the lead view and the member view side by side; they are the same artifact. Second, the subject rule: what crosses is about the work, and person-shaped material is withheld before the model sees it and checked again after. Third, the preview: each member can watch exactly what of theirs would feed the Mirror and pull it. A pilot exists so a real team can run all three checks on its own material.

What happens to our data if we cancel?

The team's aggregate history is archived, each member's private space remains theirs, and members leaving take their own history with them. Conversations are never sold and never used to train foundation models.

Why should we trust a product with no customers?

Do not trust it; verify it. Everything load-bearing in the pitch is a structural property you can check in the product: the symmetric render, the subject rule running in code, the contribution preview. What a young product honestly cannot offer is track record, which is why the pilot is flat-fee, capped at 15 people, and credits to the first year: the cost of finding out is bounded and stated up front.

Is Teams anonymous?

We do not promise anonymity, because on a small team that is not a promise anyone can keep. The claim is narrower and always true: the Team Mirror carries the team's patterns with no names attached, and anything beyond that moves only because a member sent it to their lead, with their name on it or without, exactly as they chose. A lead on a team of three may still guess where something came from. Nothing about a person crossed to help them.

---

Run a Team Mirror with your team: $990 flat, up to 15 people, and the fee credits to your first year. See how it works. Related: how Teams works and what a Glass Wall is.