Effective November 2026. Reviewed every six months.
Why we publish this
WHO Digital Strategy uses AI tools every day. We use them to analyze ad accounts, research markets, plan content and draft reports. Our clients trust us with their accounts, their data and their reputation, so they should know which tools we use, what we put into them and who checks the results. This page is part of our transparency around usage. If anything here is unclear, ask us and we will explain it.
How we use AI
AI speeds up our work. Every decision on a client account is still made by a person on our team. We use AI for:
- Campaign analysis, such as reviewing search terms, spotting unusual spend and comparing performance over time.
- Reporting, including first drafts of monthly and quarterly summaries.
- Research into competitors, audiences, keywords and how a brand appears in AI search results.
- Content planning and first drafts of articles, page copy and ad copy.
- Site and landing page audits.
- Testing a strategy from several angles before we recommend it.
- Drafting emails and meeting notes.
- Automating routine checks, such as daily budget alerts.
A member of our team reviews every AI-assisted output before it reaches a client, goes live on a website or changes an ad account.
Telling you when we use AI
AI is part of how we deliver every engagement, and we tell clients that at the start. If you want to know whether AI helped with a specific deliverable, ask and we will tell you which tools were involved and how.
You can opt out. If you would prefer that your data stay out of AI tools, tell us and we will keep it out. Opting out can change the scope and timelines of the work, because some analysis takes longer by hand, and we will tell you which parts are affected before anything changes. Opting out does not switch off the AI built into Google Ads or Meta Ads, such as automated bidding. Those features are part of the platforms themselves. If you want them off, we can run campaigns with manual bidding and will explain the trade-off.
Approved tools
We use paid, vetted tools for client work. We do not use free consumer AI tools with client information.
- Claude (Anthropic) is the main tool we use to work on client data. Every WHO Digital Strategy account runs on a paid plan with the setting that lets Anthropic train its models on our conversations switched off.
- Granola and Plaud take meeting notes. Granola covers our internal meetings and Plaud covers client calls and conferences.
- Google Ads, Meta Ads and Google Analytics 4 include built-in AI features, such as Smart Bidding, Performance Max, responsive search ads and GA4 insights. We use them where they suit the campaign.
- The ChatGPT ads platform (OpenAI) runs contextual and location-targeted campaigns for clients who choose that channel.
- Ahrefs, DataForSEO and Relational SEO provide search and AI visibility data.
- Gemini image generation and Gamma create visuals and presentation drafts. We do not put client personal data or confidential information into them.
Before we adopt a new tool for client work, we review how it stores data, whether it trains on what we enter and who can access it. We update this list when the tools change.
What goes into AI tools
We use AI tools with:
- Your business name, website and public contact details.
- Campaign performance data, search term reports and account structure.
- Analytics and Search Console data.
- Public content, such as your website pages, ads and listings.
- Notes from our meetings with you, once you have agreed to recording (see below).
What never goes into AI tools
- Passwords, API keys and other login details.
- Bank, card and payment information.
- Personal data about your guests, donors, patients or customers, unless we have removed names and contact details first. We remove that data before any analysis starts.
- Confidential business information, such as contracts, financials and unreleased plans, unless you approve it for a specific task.
Our test: if you would not want it read aloud in a room of strangers, it stays out of AI tools.
Meeting recordings
We ask for consent to record client meetings when you sign with us. That consent covers our calls with you, and you can withdraw it at any time by telling us. After that, we take notes by hand. Recordings and notes are used only for your account.
Human oversight
A person on our team makes every decision. Before anything reaches you or your account, a member of our team:
- Reviews every AI output.
- Checks figures against the source platform, such as Google Ads, GA4 or Search Console.
- Decides budgets, bid strategies, campaign structure and content direction.
- Writes or edits every client message before it is sent.
Some platform features, such as Smart Bidding, change bids automatically within the goals we set. For those, we set the targets, budgets and limits, monitor results and step in when performance moves outside them. We also run automated spend alerts on the accounts we manage.
Accuracy
- We check AI-generated facts before we present them to you.
- When we are unsure of something, we say so.
- We flag data that is incomplete or not yet verified.
- We never invent data, screenshots, results or reviews.
- Published content is edited by a person and checked against named sources.
Who owns the work
Once you have paid for a deliverable, you own it. This covers content, reports, ad copy, schema and any other work we produce for your account, whether or not AI helped create it. WHO Digital Strategy keeps ownership of its own methods, skills, templates and internal systems, which we use across all our engagements.
Data storage and security
Client data is kept in private, access-controlled repositories. Each client has a separate folder, and we never combine one client’s data with another’s. Access is limited to the WHO Digital Strategy team, and we use two-factor authentication on every platform that supports it.
Laws we follow
We serve clients in the United States and Europe.
- GDPR. Personal data about people in the EU and UK does not go into AI tools unless it has been anonymized first.
- Texas Responsible AI Governance Act. We are based in Texas and follow the act, which took effect on January 1, 2026. We do not use AI to manipulate people, to discriminate unlawfully or to create deceptive likenesses of real people.
- US state privacy laws. We handle consumer data in line with the privacy laws of the states where our clients operate.
Keeping this policy current
We review this policy at least every six months. We also review it sooner when:
- We adopt a new AI tool.
- A vendor such as Anthropic, Google, Meta or OpenAI changes how it handles data.
- A client raises a concern.
- New AI or privacy law takes effect in Texas, the US or the EU.
Questions
If you have any questions about how we use AI, email us at [email protected].
Last updated: November 2026.
