# Tideproof

> Tideproof is an educational web app for AI-era career and wealth planning by Aunova Technologies Ltd. (Dubai International Financial Centre, UAE). The AI tide is rising. Tideproof scans how exposed your work is, maps four plausible futures, and gives you a block-by-block plan for skills, income and assets that hold up in all of them. The free parts work today; Pro and Teams are launching soon and nothing is sold yet.

Free today: the AI exposure scan, scored by hai, the site's decision model; 4 futures (Slow Burn, Abundance Dividend, Two-Speed World, Owners' Economy), set by how fast AI capability arrives and how widely its gains are shared; 6 building blocks for a plan (Skills AI amplifies, Ownership, Income streams, Trust & network, Place & cost base, Health & energy); an asset lens across 6 asset classes (Home & real estate, Stocks & equity, AI infrastructure, Cash & bonds, Land & scarce assets, Yourself); and an AI toolkit (NanoGPT, Coursera, beehiiv, Framer, ElevenLabs).

Also free, for founders and CTOs: an AI-visibility check of your product's website, run by Aunova, the company behind Tideproof, requested through the site and emailed to you. It returns a readiness grade across technical extractability and on-page signals, plus a spot check of whether an AI assistant names you for your category.

Planned launch pricing, launching soon, nothing is sold yet: Pro $12 per month or $96 per year; Teams $6 per seat per month, from 25 seats. There is no checkout: every paid plan only collects an email for the launch waitlist.

Education, not financial, investment, legal or tax advice. Sources cited on the site: Stanford Digital Economy Lab, "Canaries in the Coal Mine?" (Aug 2026 update); World Economic Forum, Future of Jobs Report 2025; IMF, "AI Will Transform the Global Economy. Let's Make Sure It Benefits Humanity" (Jan 2024). Contact: info@tideproof.org.

## Pages

- [Tideproof](https://tideproof.org/index.md): the full landing page as markdown, with the futures, the building blocks, the asset lens, the AI toolkit, planned pricing, 8 questions and answers, sources and the disclaimer
- [Privacy](https://tideproof.org/privacy.md): draft privacy notice (What we process, Your rights)
- [Terms](https://tideproof.org/terms.md): draft terms of use (No advice, Scores, Referral links, Early access)

## For agents

The exposure scan is a public JSON endpoint, rate-limited per IP. robots.txt keeps crawlers out of `/api/`: call it only when a person asks you to scan their work. What happens to the text you send, in the same words the site gives the person: Your text is sent over an encrypted connection to hai, our scoring model, and scored on five independent dimensions. We do not store it, and it is never attached to a waitlist signup: only the resulting score, band, sector and suggested blocks are. Running hai means sending your text to our AI model provider, which processes it for us under its data processing terms, may be located outside the UAE, and does not use it to train models.

```http
POST https://tideproof.org/api/scan
Content-Type: application/json

{"description": "<job description, 12+ chars>"}
```

`description` is trimmed of control characters, has its whitespace collapsed, and is truncated to 600 characters before scoring, so send the 600 that matter.

A successful response (HTTP 200) is JSON with `"ok": true` and:

- `score`: exposure to AI, higher means more exposed. Reported on a 0 to 100 scale but clamped to 3..97 inclusive, so 0 and 100 never occur.
- `band`: `Low` (under 30), `Moderate` (30 to 49), `High` (50 to 69), `Very high` (70 and above).
- `sector`: one of `tech`, `creative`, `finance`, `legal`, `health`, `trades`, `education`, `sales`, `admin`, `leadership`, `hospitality`, `other`.
- `dimensions`: the scores behind the result, each `{ key, label, value, hint }` with `value` from 0 to 1.
- `plan`: the building blocks to start with, each `{ block, name, why }`; `block` is one of `skills`, `ownership`, `income`, `trust`, `place`, `health`.

Errors are JSON `{"ok": false, "error": "<message>"}`: 400 when the body is not JSON, is longer than 4,000 characters, or `description` is missing, is not a string, or is shorter than 12 characters; 422 when the text does not describe work; 429 when over the rate limit (retry after the `retry-after` seconds); 403 for browser requests from another origin.