Privacy Policy
Last updated: 9 October 2026 · Effective: 9 October 2026
Questions about this policy? Email mohamedsidiyot@gmail.com.
1. Who we are
Ikigai ADE is the trading name of Mohamed Sidiyot, an individual trading as a sole proprietor in South Africa. Ikigai ADE is not a registered company; there is no separate juristic entity, and Mohamed Sidiyot is personally the contracting party.
Mohamed Sidiyot, a sole proprietor trading as Ikigai ADE is the responsible party (the "controller" under the GDPR) for personal information collected through the Ikigai ADE platform and website.
Information Officer: Mohamed Sidiyot
Email: mohamedsidiyot@gmail.com
Tel: +27 68 695 9713
South Africa
Disclosure on our own compliance status: POPIA section 55 read with Regulation 4 requires an Information Officer to be registered with the Information Regulator. Our registration is not yet complete. We are telling you this rather than implying a compliance status we do not hold. It does not reduce any of your rights below, and it does not affect your right to complain to the Regulator.
We do not currently publish a physical business address. Section 43 of the Electronic Communications and Transactions Act expects an online supplier to do so, and we have not yet met that requirement. Reach us at the email and phone number above.
2. Information we collect
Account information. Name, email address and authentication data you provide at sign-up. Authentication is handled by Supabase.
Information you volunteer in forms. On our waitlist and contact forms: your name, email, and optionally a phone number, company name and website. Optional fields are labelled optional and you can leave them blank.
Content you put into the product. Prompts, uploaded documents, CRM contacts, brand material and anything else you add to your workspace or your Brain. This is yours. We process it to run the product for you.
Conversation memory. Context from your conversations with S.I.D and the agent team, stored against your account so a conversation can continue where it left off.
Usage data. Pages visited, features used, credits consumed, and which model served a request. We use this to run billing and to see where the product is failing people.
Homepage visit data, only with your OK. Our public homepage asks whether you accept analytics. If you do, it records under a random id which sections you reach, how far you scroll, what you click, where the visit came from, the kind of device, and whether you start or finish the waitlist form. No name, email or IP address is stored with it, it stays in our own database, and it is deleted 180 days after your last visit. We use it to see where the page loses people, to test two versions of a heading, and to show you a version of the page that fits your last visit. Choose "Essential only" and nothing is recorded under any id. Detail is in the Cookie Policy.
Technical and error data. When a page crashes, your browser sends us the error message, a stack trace, the page you were on, and your IP address is visible to us in that request and in our hosting logs. An IP address is personal information, so we are naming it here rather than hiding it under "technical data". We use it to diagnose faults and to rate-limit abuse.
Payment data. Card details are entered with Paystack and never reach our servers. We receive your name, email, billing country, the last four digits of the card, and whether the payment succeeded.
Third-party contact data you give us. If you upload prospects or contacts into the CRM or outreach tools, you are giving us other people's personal information. You remain the responsible party for that data; we process it on your instruction. You must have a lawful basis for it, and our Terms require you to.
3. Why we are allowed to process it
POPIA section 11 and GDPR Article 6 both require a specific ground for every purpose. Ours, purpose by purpose:
| What we do with it | Ground we rely on |
|---|---|
| Create and run your account; deliver the product you subscribed to | Performance of a contract with you |
| Take payment, issue invoices and calculate credits | Performance of a contract; compliance with tax law |
| Send transactional email (receipts, password resets, service notices) | Performance of a contract |
| Diagnose crashes, keep logs, rate-limit abuse, investigate security incidents | Legitimate interest in a working and secure service |
| Keep records for tax and accounting | Legal obligation (Tax Administration Act) |
| Waitlist and marketing email you asked for | Your consent, withdrawable at any time |
| Record what visitors do on our homepage, under a random id | Your consent, given in the cookie choice and withdrawable at any time |
| Cold outreach to business prospects (email, SMS, voice, Telegram) | Legitimate interest, subject to POPIA s.69 and your right to object at first contact |
| Improve the product using aggregated, de-identified usage patterns | Legitimate interest; the data is no longer personal information |
Where we rely on consent, you can withdraw it at any time and we stop. Where we rely on legitimate interest, you can object and we will either stop or explain why our interest overrides your objection. Withdrawing consent does not undo processing that was lawful before you withdrew it.
We do not sell your personal information, and we do not use it to build advertising profiles. We may create and share aggregated, de-identified statistics that cannot reasonably be traced back to you or any individual. That is not personal information under POPIA or the GDPR. We do not try to re-identify it and we require anyone who receives it not to either.
4. AI processing and model training
The product works by sending your prompt and the context it needs to a large language model. That means the content you submit leaves our infrastructure and is processed by the AI provider serving that model. Which providers, and in which countries, is set out in sections 6 and 7.
- We only send what the request needs: the prompt, the relevant context, and your workspace content where you have asked an agent to use it.
- We contract with providers on terms that do not permit them to train their models on your content. Providers may retain it briefly for abuse monitoring, typically up to 30 days. The per-provider position is on the Sub-processors page.
- If you enable Bring Your Own Key, requests go to your own provider account under your own agreement with that provider, and their terms govern retention, not ours.
- We do not train our own models on your workspace content. Where improvements are shared across customers, what moves is the generalised technique, never your data. See the shared-learning disclosure.
Do not paste anything into the product that you are not willing to have processed by a third-party AI provider outside South Africa. If that is a problem for your data, use Bring Your Own Key, or ask us to route around a specific provider under section 9.
5. Automated decisions and profiling
Some features score or rank people automatically: lead scoring, ICP fit, churn risk and similar. POPIA section 71 and GDPR Article 22 restrict decisions made purely by automated means that have a legal or similarly significant effect on someone.
- These outputs are recommendations for a human to act on. Nothing in the product automatically refuses anyone a service, a price or a contract.
- If you use our scoring to make decisions about other people, that is your processing and your responsibility under Article 22 and section 71, not ours.
- You can ask us how a score about you was produced, and ask for it to be reviewed by a person.
6. Who else processes your data
POPIA section 21 and GDPR Article 28 require us to name our operators (sub-processors). The full, dated and authoritative list lives on the Sub-processors page, which we update before a change takes effect. Summarised:
| Category | Who | What they process |
|---|---|---|
| Hosting and network | Vercel, Cloudflare | Request metadata, IP addresses, logs |
| Database, auth, storage | Supabase | Account, workspace content, CRM, billing records |
| Payments | Paystack | Name, email, billing address, tokenised card |
| AI model providers | Anthropic, OpenAI, Google, Groq, NVIDIA, xAI, Moonshot AI, DeepSeek, Zhipu AI, Alibaba | Your in-flight prompt and its context |
| Resend | Recipient address and message body | |
| SMS and voice | Twilio | Phone numbers, message bodies, call audio |
| Text to speech | ElevenLabs | The text to be spoken |
| Messaging | Telegram | Chat identifiers and message content, if you connect it |
| Lead capture | Google (Apps Script and Sheets) | Waitlist and contact submissions: name, email, phone, country, company, website |
| Logging and uptime | Better Stack | Request metadata, stack traces |
| Source control | GitHub | Code only, no customer data |
We require every one of them to process data only on our instructions, to keep it secure, and to be bound by a data-processing agreement. We will tell active customers at least 30 days before adding a new sub-processor.
7. Where your data goes (cross-border transfers)
Your personal information leaves South Africa. It has to, because the infrastructure and the AI models we use are hosted abroad. POPIA section 72 and Chapter V of the GDPR both require us to say so and to name the basis for it, so here it is plainly.
| Destination | Why it goes there | Safeguard |
|---|---|---|
| United States | Hosting, database, most AI model providers, email, SMS, logging | Contractual terms in each provider DPA imposing protection comparable to POPIA (s.72(1)(a)); GDPR Standard Contractual Clauses |
| European Union | CDN edge, logging (Germany), some Google regions | Adequate protection under POPIA s.72; within the EEA for GDPR |
| Nigeria and South Africa | Payment processing (Paystack) | Provider DPA; Nigeria has the NDPA 2023 |
| Singapore | Alibaba DashScope, only if you select Qwen with your own key | Provider DPA; PDPA applies locally |
| China | Moonshot AI (Kimi), DeepSeek and Zhipu AI (GLM) when those models are selected | Provider terms only. See the warning below. |
| Global CDN edge | Static assets and routing | Request metadata only, no workspace content |
Please read this if you are in the EU or the UK. Three of the model providers we can route to are based in China. There is no adequacy decision for China under the GDPR, and we cannot honestly claim that contractual clauses alone give you protection equivalent to EU law there. If you are subject to the GDPR, or your data is sensitive, either avoid selecting those models, or tell us to exclude them from your account under section 9 and we will. GLM and Qwen have no managed lane at all: they only run on your own key.
8. How long we keep it
| Data | How long we keep it |
|---|---|
| Account and profile | While your account is active, then deleted within 30 days of account deletion |
| Workspace content, Brain, CRM | While your account is active; deleted with the account |
| Conversation memory | While your account is active, or until you clear it |
| Prompt content held by AI providers | Per provider, typically up to 30 days for abuse review, then deleted |
| Billing and tax records | 5 years after the tax year, as the Tax Administration Act requires. We cannot delete these on request. |
| Error logs and security logs | 90 days |
| Waitlist entries | Until you ask us to remove you, or 24 months of no contact |
| Homepage visit records (random id, no name or address) | 180 days after the last visit |
| Aggregated, de-identified analytics | Indefinitely. It is no longer personal information. |
9. Your rights
Whatever your jurisdiction, you can ask us to:
- Access a copy of the personal information we hold about you.
- Correct anything inaccurate or incomplete.
- Delete it, except where we are legally required to keep it (see section 8).
- Export it in a machine-readable format.
- Object to processing based on legitimate interest, including outreach.
- Withdraw consent where consent is the ground we rely on.
- Restrict processing while a dispute about accuracy or lawfulness is resolved.
- Exclude a specific sub-processor, in particular an AI provider or a country. Where it is technically possible we will reroute you, and where it is not we will tell you so.
- Not be discriminated against for exercising any of these. Under the CCPA we will not degrade your service or change your price because you asked.
Email mohamedsidiyot@gmail.com. We respond within 30 days. We may ask you to confirm your identity first, because handing your data to someone impersonating you would be worse than a slow reply. Asking costs nothing.
11. Security
- HTTPS everywhere, with strict transport security.
- Postgres Row Level Security so one customer's rows cannot be read by another.
- Your own provider API keys are stored encrypted with envelope encryption, never in plain text.
- Rate limiting and abuse controls on public endpoints.
- Access to production data is limited to the Information Officer.
No system is perfectly secure and we will not pretend otherwise. If a breach affects your personal information, POPIA section 22 requires us to notify you and the Information Regulator as soon as reasonably possible; under the GDPR we will notify the relevant authority within 72 hours. Report a suspected vulnerability to mohamedsidiyot@gmail.com.
12. Children
Ikigai ADE is a business product and is not directed at children. You must be 18 or older to hold an account. We do not knowingly collect personal information from anyone under 18. If you believe a minor has created an account, email mohamedsidiyot@gmail.com and we will delete it.
13. Changes to this policy
We may update this policy. If a change is material, we will email registered users at least 14 days before it takes effect, and we will change the "last updated" date above. We will not apply a materially worse policy to data we already hold without telling you first.
14. Complaints
Please raise it with us first at mohamedsidiyot@gmail.com. You do not have to, and you are entitled to go straight to a regulator.
South Africa. Information Regulator (South Africa): inforegulator.org.za
Email: POPIAComplaints@inforegulator.org.za
EU and UK. You may complain to your national data protection authority, or to the Information Commissioner's Office in the UK.
15. Accounts you connect, including Google
You can connect outside accounts in Settings › Models & keys, or from the section that uses them. Each one is connected only when you sign in to it yourself and agree on that service's own screen, and each is used only for the feature you connected it for. You can disconnect any of them at any time in the same place.
Google. What each Google permission is used for:
- Google Drive (only the files you choose). When you pick files in Google's own picker, we copy those files into your Studio library so you can edit them, post them or build a website from them. We cannot see the rest of your Drive.
- Google Calendar (your events). When a call is booked, we add it to your calendar with a Google Meet link and invite the other person; we move or remove it when the booking changes. We read when you are busy so your booking page does not offer those times, and we show today's meetings in Mission Control. We keep the event's id and Meet link with the booking; we do not copy your other events.
- Gmail (read and send). We show your conversations in Inbox and send the replies and outreach emails you write or approve, from your own address. Your mail is read from Google when you open it and is not stored by us; for outreach we keep the id of the conversation we sent so we can tell when the person replied.
- YouTube. We upload the videos you ask Social to publish and read your channel's details and results.
- Search Console, Analytics, Business Profile and Google Ads (read only). We read your search, visitor, review and campaign numbers when you run a report in Intelligences, and show them in that report.
How Google data is handled.
- The sign-in tokens Google gives us are stored encrypted, are opened only on our servers when a feature you use needs them, and are never shown to anyone.
- We do not sell Google user data, use it for advertising, or use it to build profiles of you or anyone else.
- We do not use Google user data to develop, improve or train any AI or machine learning model, ours or anyone else's. Where you ask an AI feature to work on it (for example, a report that reads your Search Console numbers, or a website built from a Google Doc you picked), only what that request needs is sent to the AI provider, under terms that do not allow the provider to train on it (section 4). Your Gmail messages are never sent to an AI provider.
- Nobody at Ikigai ADE reads your Google data, unless you ask us to for support, it is needed for security or to comply with the law.
- We share Google user data only with the processors listed in section 6 that are needed to run the feature, or where the law requires it.
- Disconnecting a Google account in Settings › Models & keys deletes its tokens from our systems and asks Google to end our access. You can also remove Ikigai ADE at myaccount.google.com/permissions. Files you brought into your library and bookings you made stay until you delete them or close your account (section 8).
Ikigai ADE's use and transfer to any other app of information received from Google APIs will adhere to the Google API Services User Data Policy, including the Limited Use requirements.
Microsoft Outlook, Meta, Xero, GitHub and Supabase. The same rules apply: Outlook is used to show your mail in Inbox and send what you write or approve; Meta Ads and Xero are read only, for your reports and your books; GitHub and Supabase are used to save and link the projects you build. Tokens are stored encrypted, nothing is sold or used for advertising or model training, and disconnecting deletes the tokens here.