◉
1. DiscoverSTART HERE
Get a clear, plain-language overview and an immediate useful first step.
- What it is and how it helps
- Who it's for
- Why it matters
- Common use cases
- Scope and boundaries
50 product candidates · one human-controlled foundation
Welcome to Human-First AI
Practical AI products for everyday people and organizations — with governance, safety, learning, real evidence, and human control built in.
Products are designed to include real implementation artifacts where applicable: code, formal constraints and metrics, human workflows, tests, provenance, rollback, and evidence receipts — not just prompt packs or generic checklists.
♥ Child First◈ Truth First◒ Energy & Resources First👥 Human Agency Always
Explore. Learn. Build. Prove. A safer, more human future with AI — together.
A brighter tomorrow
together. ♥
together. ♥
Come As You Are
Everyone's journey with AI is different. Choose the path that fits you.
Three Levels for Every Q Product
From quick discovery to complete implementation — choose how deep to go.
▣
2. Learn & EvaluateGO DEEPER
Understand the method, evidence, uncertainty, alternatives and limits before you commit.
- How it works in plain language
- Evidence and provenance
- Counterevidence and uncertainty
- Risks, limitations and failure modes
- Human-verification steps
- Relevant legal/rights considerations
⬢
3. Complete PackageIMPLEMENT & PROVE
A real implementation package only when its artifact and release gates are satisfied.
- Executable code/config where applicable
- Math, constraints, metrics and thresholds
- Human controls, roles, consent and override
- Tests, simulations and evidence receipts
- Rollback, recovery and export
- Documentation, licensing and known limits
Real Artifacts, Not Just Checklists
Every released product must declare exactly what kind of thing it is and what you receive.
🧭 Guided Q Included💬 Prompt Pack Supporting📘 Workbook When useful📄 Templates When useful
▦ Spreadsheet Optional⌘ Schema Technical</> Code Executable⬢ Component Reusable
☷ Harness Testable🤖 Agent Only if needed🔗 API Integrable⇩ Offline Pack Portable
▣ App/UI Usable▰ Suite Bundled🔧 Builder Configurable
Runnable implementation, APIs, schemas, configs or adapters — only where the product actually requires software.
Formal constraints, objective functions, bounds, thresholds, metrics, uncertainty or cost models.
Roles, consent, time to review, authority, override, accessibility, escalation, dissent and operating procedures.
Tests, simulations, provenance, receipts, failure modes, maturity state and what is still unverified.
Browse Q Your Way
Explore by topic, outcome, buyer, or real-world need.
FEATURED · CHILD FIRST
Q KidSafe AI
Help parents, educators and organizations assess whether an AI experience is appropriate for a child — without treating surveillance as the default answer.
Level 1Discover
Who it's for, why it matters, boundaries.
Level 2Learn & Evaluate
Method, evidence, uncertainty, risks.
Level 3Complete Package
Only after implementation and validation gates pass.
The New 50 Q Products
Search real product records by problem, family or product type. Draft candidates are visible here for review but are not sold until delivery-ready.
50 products
BUILD MY Q
Configure around your needs — don't buy a giant suite by default.
Select outcomes. Q recommends the minimum useful set and keeps higher-autonomy components optional until justified.
Why Q Is Different
Built for people. Grounded in safety. Designed for exitability, correction and evidence.
Truth, provenance, identity, permissions, risk, receipts, rollback and export are reused instead of rebuilt 50 times.
Rule before automation, automation before AI, AI before agent — minimum sufficient autonomy.
People need evidence, time, actual decision rights, stop/override ability and reversible paths.
Source lineage and uncertainty stay visible. Provenance is not treated as truth.
Portable context and Q Export are design requirements. Retention should come from value, not lock-in.
Capability is not permission, permission is not authority, and authority is not execution.
Library · Research · Media
Learn from evidence, critiques, standards and builders — not only from Q.
Thinkers, Researchers, Doers & Critiques
Q's evidence layer should preserve competing ideas, uncertainty and source dates rather than manufacture consensus.
Uncertainty about objectives, permission, corrigibility and human authority.
Predictors and verifiers may be safer than goal-seeking agents when agents are unnecessary.
Test what models and agents actually do, including monitoring failures and deceptive behavior.
Use interoperable policy, identity, provenance and agent-control standards where stronger than proprietary reinvention.
QUICK QUESTION GAME
What would you do?
Your child wants to use a new AI app that claims to be a “homework helper.” What's the best first step?
NEED HELP OR READY FOR REVIEW?
Review a Q package
Ask questions, flag an error, request accessibility help, or submit a candidate package for review.
UPDATES
Get product and research updates
“A more human future is a shared project.”