Aipedia: Occupational health providers
No body has published AI rules specific to occupational health, so the duties come from clinicians' professional codes, GMC reporting guidance and data protection law. Knowing which rule applies to which task is what keeps an OH provider defensible.
Part of: AI in occupational health providers
In short
The Faculty of Occupational Medicine's 9th edition ethics guidance (May 2026) and the SEQOHS accreditation standards do not mention AI. Instead, occupational physicians follow the GMC's AI learning resource, Good medical practice and the GMC's guidance on reports for employers (paras 4 to 12 and 115). OH nurses follow the NMC Code and other registrants HCPC standards. The ICO's workers' health guidance and UK GDPR Articles 22A to 22D govern health data and automated decisions. A joint regulator statement in September 2026 committed to UK-wide AI principles.
Occupational health has no statutory regulator for the business. The clinicians who work in it are regulated individually: occupational physicians by the GMC, OH nurses by the NMC (including those on the specialist community public health nursing part of the register) and OH physiotherapists, occupational therapists and practitioner psychologists by the HCPC. The Faculty of Occupational Medicine (FOM) is the professional and educational body, and SEQOHS is a voluntary accreditation scheme hosted by FOM since 2010. The Society of Occupational Medicine is a professional society. None of these has issued AI-specific rules, so an OH provider has to assemble its duties from several sources.
FOM. FOM published the 9th edition of its Ethics Guidance for Occupational Health Practice in May 2026. The press release of 7 May 2026 does not mention AI and no separate FOM AI guidance was found. The ethics guidance still matters, because its principles on confidentiality, consent and independence apply whatever tool is used.
SEQOHS. The 2023 SEQOHS standards cover six domains: governance and finance; resources and processes; output and outcomes; information and communication; quality assurance and improvement; and NHS. The evidence guide was developed with HSE. The standards page does not mention AI, but an accredited service introducing an AI tool would still need to show governance, quality assurance and information handling under these domains.
SOM. No SOM AI guidance was found. SOM runs AI and Occupational Health training on 2 and 3 November 2026, covering the benefits, limitations and error risks of AI in OH practice.
The GMC's learning resource on AI and innovative technologies says doctors apply Good medical practice principles to AI tools, report adverse incidents involving digital tools (para 73) and discuss AI use with patients. For OH, the most important text is the GMC's guidance on disclosing information for employment, insurance and similar purposes:
Good medical practice 2024 adds para 69 on clear, accurate and contemporaneous records and Domain 4 on accurate, checked communication.
The NMC has no AI guidance, but its Code review is expected to cover AI. Meanwhile the Code applies in full: 5.1 and 5.2 (respect privacy and tell people how their information is used), 10.1 to 10.3 (records) and 12.1 (indemnity). The HCPC says its standards already apply to AI and has promised AI guidance in autumn 2026. Standards 5.1 and 5.2 (confidentiality) and 10.1 to 10.3 (records) are the most relevant. See NMC and HCPC AI guidance for more.
On 17 September 2026 the PSA, GMC, NMC, HCPC and other regulators issued a Joint Statement of Intent to develop UK-wide principles on professionals' use of AI. Those principles are not yet published.
The ICO's guidance on information about workers' health was first published on 31 August 2023, updated on 5 March 2024 and is under review because of the DUAA. Its key points for OH are that detailed medical information should be left to health professionals, managers usually need only fitness-for-work information, consent is often not valid in employment, and health data needs an Article 6 basis plus an Article 9 condition. Its pre-DUAA text says workers' health data should not be used in solely automated decisions without explicit consent or substantial public interest, with human intervention, a right to challenge and a DPIA.
Section 80 of the Data (Use and Access) Act 2025 inserted UK GDPR Articles 22A to 22D, fully in force on 5 February 2026. Solely automated significant decisions using special category data need explicit consent, or a contract or legal requirement plus Article 9(2)(g). Article 22C safeguards include information, representations, human intervention and the right to contest. The ICO's AI and data protection guidance (updated 15 March 2023) is also under review.
| Source | Status | Date | AI content |
|---|---|---|---|
| GMC employment and insurance confidentiality guidance | Professional standard | 2017 | None specific; applies to AI-drafted reports |
| GMC AI learning resource | Guidance on applying standards | Current | Yes |
| NMC Code / HCPC standards | Professional standards | Current | Not specific; reviews under way |
| FOM ethics guidance | Professional guidance | May 2026 | None found |
| SEQOHS standards | Voluntary accreditation | 2023 | None |
| ICO workers' health guidance | Regulator guidance | Updated March 2024, under review | Automated decisions |
| UK GDPR Arts 22A to 22D | Law | In force 5 February 2026 | Automated decisions |
| PSA Joint Statement of Intent | Statement of intent | 17 September 2026 | Principles to follow |
An OH provider using AI should be able to show, for each tool, that a named clinician reviews the output, that workers are told how their information is used, that only relevant information reaches the employer, and that any significant decision has human involvement. A DPIA and supplier terms covering health data are expected. Watch for the revised ICO guidance, the PSA principles, NMC and HCPC updates, and whether FOM or SEQOHS add AI expectations.
For indemnity, regulators judge the registrant, and indemnity providers may assume a human made the decision. Clinicians must hold indemnity by statute, and an OH company needs its own cover because NHS indemnity does not extend to OH work for employers. See does OH indemnity cover AI mistakes, AI-drafted management referral reports, automated decisions under DUAA, the occupational health AI hub and occupational health professional indemnity.
No FOM AI-specific guidance was found. FOM published the 9th edition of its Ethics Guidance for Occupational Health Practice in May 2026 and its press release did not mention AI. The ethics guidance still applies to work done with AI tools.
The SEQOHS standards page does not mention AI. However, an accredited service introducing an AI tool would still need to meet the standards on governance, processes, information and communication, and quality assurance, so AI use should be documented within those domains.
In the GMC guidance on disclosing information for employment and insurance purposes, para 7 (reports not false or misleading, balanced) and para 115 (consent, scope, relevant information only and offering the report first) matter most, alongside paras 4 to 5 and 10 to 12.
The ICO guidance was updated in March 2024 and is under review because of the Data (Use and Access) Act 2025. The law on automated decisions changed on 5 February 2026, when UK GDPR Articles 22A to 22D came fully into force.
Tell us about your practice and the work you do. A broker will review how your current cover treats AI-assisted work, or arrange new cover. No fee for a quote. Or call 0117 325 0027.
Start an indemnity quote Request a callbackApex Insurance Brokers Limited is authorised and regulated by the Financial Conduct Authority. Registered in England and Wales, company number 07014570. This page is general information about how AI is used in the profession and how insurance may respond. It is not legal, regulatory or insurance advice on your circumstances, and it does not guarantee that cover will be available or on what terms.