The coaching case for AI disclosure
The default reaction from solo and small-firm coaches has been: "The EU AI Act is a Big Tech rule. I'm one ICF-credentialed coach with twelve clients on a Calendly link." Article 50 doesn't carve any of that out. It applies to providers and deployers of AI systems whose output is "made available to the public" — and a session-recap email, an AI-scored intake form, or a GPT-powered chatbot embedded on a Kajabi sales page all meet that definition. There is no minimum-headcount threshold, no "but it's 1:1" carve-out, and no hobbyist exemption. The geographic gate is whether any of your clients or newsletter subscribers is in the EU, which for any public-by-default funnel is essentially always yes.
The penalty headline is €15 million or 3% of global turnover, whichever is higher. Solo coaches won't be regulators' first target, but the practical risk arrives faster: an ICF or EMCC ethics complaint citing hidden AI use, a refund claim from a client who feels the relationship was misrepresented, or platform-level penalties as Kajabi, Teachable, and ConvertKit roll out attestation requirements through 2026. The cheapest insurance is a per-asset label on the AI elements that actually appear in your practice — nothing more, nothing less. This generator outputs that label in three formats simultaneously.
The five AI elements coaches actually publish
Most coaches don't need to label everything. The presets map to the five real cases:
- AI assessments and intake. The most common case. AI-generated values inventories, AI-scored 360s, GPT-drafted intake questions. The generator emits a top-of-form consent line and a JSON-LD
Questionnaireblock where the AI iscreatorand you arereviewer. - AI session summaries. Otter.ai, Fathom, Granola, Read.ai. Article 50 calls this "synthetic text." The generator emits a top-of-email tag, an email-signature attribution block, and a JSON-LD
EmailMessagesnippet for any web-archived recap. - Coaching chatbots. The highest-disclosure case because chatbots get their own Article 50 clause for "systems that interact with natural persons." The generator outputs an opening-message disclosure, a persistent footer line, and JSON-LD
ChatAction— the three placements regulators have signaled they'll check first. - AI-written workbooks. Where AI drafted significant client-facing content. Two preset bands: "AI-assisted" (you wrote the spec, AI filled examples) and "AI-generated" (AI wrote the whole workbook). Pick the firmer one if the contribution split is genuinely close.
- AI voice in audio deliverables. ElevenLabs or PlayHT in your cloned voice for evening reflections, walking-meditation prompts, or sales-page audio. The generator emits a 4-second spoken intro and an RSS feed block for any podcast or audio-coaching distribution.
Where to place the disclosure on a coaching deliverable
The legal text says "clearly and distinguishably perceptible at the latest at the time of the first interaction or exposure." For coaches, that means three placements that survive how clients actually consume your work:
- Top of the email or PDF. Above the first piece of content, not in the footer. A small badge plus "AI-drafted, reviewed by [your name]" in plain text. This is the placement that satisfies regulators reading the email in preview pane without scrolling.
- Email-signature attribution. A two-line block under your sign-off naming the AI tool and your role as reviewer. This is the placement that survives a forwarded message — the asset travels intact.
- JSON-LD on the hosting page. If the workbook, recap archive, or chatbot lives on Kajabi, Squarespace, Webflow, or a custom site, the generator emits a JSON-LD block where the AI tool is
creatorand you arerevieweroreditor. Pastes into<head>.
JSON-LD schema: machine-readable proof of compliance
Big platforms negotiate compliance reporting with regulators directly. A solo coach can't — which makes machine-readable schema the practical equivalent of an audit log. The generator outputs schema with the AI tool named in the creator field, the human reviewer (you) named in editor or reviewer, and the model identifier carried in identifier. Both search engines and the European Commission's AI Office parse this format. If a client or platform ever asks you to demonstrate compliance, the schema is your default answer — and it travels with the page.
What this is not: ICF/EMCC ethics, GDPR, and your liability waiver
An AI disclosure tells your client an asset was generated. An ICF/EMCC ethics statement tells your client what your code of conduct is. GDPR consent tells your client how their data is processed. A liability waiver tells your client the limits of the engagement. Four different documents.
Coaches regularly try to roll these into one fine-print block. Regulators have been clear that doesn't satisfy any of them. The ICF Code of Ethics and the EMCC Global Code of Ethics both require AI-use transparency as a separate disclosure on the coaching agreement. The AI label goes on the AI element. The ethics statement goes in your engagement letter. GDPR consent goes in your intake form. The liability waiver goes in your scope-of-engagement document. The generator focuses on the first only — and does it well.
Compliance vs. theatre: what bad disclosure looks like for coaches
| Pattern | What it does | Status |
|---|---|---|
| "I sometimes use AI" one-liner in your bio | Doesn't identify which deliverable; not perceptible at the asset level | Non-compliant |
| One line in the workbook PDF footer | Reader skims past it; not at first exposure | Non-compliant |
| Email signature only, no top tag | Lost when the recap is forwarded as a quote | Borderline |
| Hidden disclosure inside a Notion toggle | Not perceptible to a reader; treated as deception | Aggravated risk |
| Top-of-email tag + signature block | Reader and forwarder both see it | Compliant |
| Top tag + signature + JSON-LD on host page | Client, forwarder, and crawler all see it | Best practice |
Workflow for a typical coaching pipeline
You don't want to revisit the generator every time you send a recap. The lighter touch is to template it. Open the generator once, build your standard variants — "AI session summary," "AI assessment," "AI-assisted workbook," "coaching chatbot" — and paste each output into a snippet manager (TextExpander, espanso, your CMS's reusable blocks, or Kajabi's saved-text feature). On send, you pick the matching snippet, fill the tool and session lines, ship. The schema block lives once in your coaching-site head; only the per-deliverable variables change. From start to send-ready takes about 60 seconds once the snippets are in place — less time than the recap itself.
Frequently asked questions
Do life and business coaches really need to disclose AI under the EU AI Act?
Yes if any client, prospect, or newsletter subscriber is in the EU. Article 50 covers AI text, image, audio, and video published to clients — including a solo coach's recap email, an AI-scored intake form, or a GPT-powered chatbot on a Kajabi page. Aug 2, 2026 deadline. Risk: fines up to €15M or 3% global turnover, plus ICF and EMCC ethical-conduct exposure.
If ChatGPT writes my coaching workbook, is that disclosable?
Yes when AI drafted a substantial part. Outline help, single-prompt rewordings, or grammar fixes don't trigger labeling. Whole workbooks or assessment rubrics do. The generator emits "AI-assisted" and "AI-generated" presets so you don't over- or under-disclose.
Will an AI label undermine the relationship?
The trust research points the other way. Clients rated coaches who openly labeled AI use as more trustworthy than coaches who used it silently and were later found out. The label frames you as the human judgment behind the deliverable, not the deliverable itself.
What about a GPT-powered "ask the coach" chatbot?
Highest-disclosure case. A chatbot is a "system that interacts with natural persons," with its own Article 50 clause. The generator emits an opening-message disclosure, a persistent footer line, and a JSON-LD ChatAction block — the three placements regulators check first.
Where do I put the disclosure on a recap email?
A one-line tag at the top above the first content block, plus an email-signature attribution under your sign-off. The generator emits both, plus a JSON-LD EmailMessage snippet for archived recaps.
Does this work with Kajabi, Notion, Calendly, ConvertKit, and Otter.ai?
Yes. The HTML banner pastes into Kajabi pages, ConvertKit broadcasts, and Notion workbooks. The signature block fits in Gmail, Outlook, and Calendly confirmation templates. JSON-LD works on Squarespace, Webflow, Kajabi, WordPress, Framer — anywhere you can edit head.