What Care Workers Expect on AI: The Careworkers' Statement of Expectations

Most guidance on AI in adult social care is written for providers, commissioners and regulators. This statement was written by care workers themselves. The Careworkers' Guidance and Statement of Expectations on the Responsible Use of AI and Particularly Generative AI in Adult Social Care sets out what frontline staff expect from the people who buy, build, regulate and manage AI tools. If you manage a care service, it is a useful way to see AI through the eyes of the people who will be asked to use it.

Where the statement comes from

According to the document, the statement was written by frontline care workers from England, Wales and Scotland who attended a roundtable on the responsible use of generative AI in adult social care. The Care Workers' Charity was the main organiser, and the roundtable was co-hosted with the Institute for Ethics in AI at the University of Oxford and Katie Thorn of the Digital Care Hub. The Charity's news item says the roundtable took place at Reuben College, Oxford, in May 2024, and that the guidance was announced on 16 September 2024. The document itself carries no publication date, and it gives two different dates for the roundtable (16 May in the header and 17 May in the text), so treat it as a 2024 statement rather than current regulatory guidance.

The statement describes itself as a first step towards putting care workers' voices at the centre of how AI is managed in social care. It opens by describing working conditions, including heavy workloads, missed breaks, unpaid training time and, for some, no sick pay. It treats AI as something that might help, but not as a replacement for care workers. In the Charity's announcement, its chief executive is quoted as saying AI should never be seen as a replacement for care workers.

What care workers expect from employers

The expectations for employers are the most practical part of the document. Care workers ask employers to:

• adopt AI and technology policies that set out clearly if, when and how AI can or should be used;
• take full responsibility for harm caused by AI where the policies and procedures were followed correctly;
• provide training and continued learning at different levels, with contact people for staff who have had less training;
• include basic AI awareness in the Care Certificate;
• provide work devices rather than expecting staff to use their own phones or computers; and
• compensate staff for any data or software costs linked to AI services.

The point about responsibility matters. The Charity's announcement says responsibility for AI use should not rest mainly on care workers. A policy that tells staff what they may and may not do, and who answers for the outcome, is the thing the statement keeps returning to.

What they expect from developers, commissioners and regulators

For developers of care-specific AI, the statement asks for products that fit the values of care, meaningful involvement of care workers in development (with fair payment for their time), honest information about limitations rather than a sales pitch, and clear information on data privacy, storage, use and safe use.

For local authorities and policy makers, it asks for responsible commissioning and usage decisions. It says AI should not be treated as a fix for every problem in the sector, that it can support but not replace care workers, and that more funding should go into training and paying care workers alongside any investment in AI.

For the regulators of the nations, it asks them to have AI training and policies within their own organisations, to communicate good practice tailored to each nation's regulations, and to train inspectors on responsible AI use. The Care Quality Commission is named only once, in the opening paragraph listing the statement's audiences. Later sections refer to regulators of the nations in general terms.

Guidance the statement gives to fellow care workers

The document also speaks to peers. It tells care workers to learn how the technology works and what its risks are before using it, especially when caring for someone, and to check whether their employer has an AI strategy, policy and procedures, and whether use is permitted. It asks them to get permission in writing, together with the terms for responsible use. It also says staff should understand data privacy, confidentiality and copyright, and should not enter personally identifiable or sensitive data into generative AI tools. Chatbots are described as tools, not people, which lack real understanding or emotion, and AI output should be checked because it can sound reliable while being inaccurate or made up.

What the statement does not cover

Reading it alongside other material, a few gaps are worth knowing about. The statement does not address the consent of people receiving care, or of their families, to the use of AI. It contains no enforcement mechanism, penalties or inspection procedure; it states expectations and asks regulators to communicate good practice. Anything about legal duties, data protection or clinical safety has to come from other sources.

Our suggestions for providers

The points below are our own practical suggestions, not part of the statement:

• Use the employer list as a checklist. Ask whether your service has a written policy saying if, when and how AI may be used, and whether every member of staff knows where to find it.
• Involve care workers early. If the people who will use a tool have helped choose it, the questions about devices, training time and data costs tend to surface before rollout rather than after.
• Pair this statement with official material. Our guide to the Department of Health and Social Care guidance on using AI in adult social care covers the government view, and our overview of AI governance in care homes looks at how providers can organise oversight. For the regulator's position, see our summary of the CQC's AI principles for care providers.

Frequently asked questions

Is this statement official government or regulator guidance?

No. It was produced by care workers at a roundtable organised by the Care Workers' Charity and partners, and it sets out their expectations. It does not create legal obligations.

Does it say care workers should not use AI?

No. It treats AI as potentially useful, but says it should support care workers and not replace them, and that staff should only use it with their employer's permission and within clear policies.

Does it say anything about pay?

Yes. It calls for better pay, payment for training time and sick pay, and for fair payment when care workers help develop AI products. It also asks local authorities to put more funding into training and paying care workers.

Who is the statement addressed to?

Employers, developers of care-specific AI, local authorities and policy makers, regulators of the nations, and care workers themselves.

Next steps for care teams

If you are planning AI awareness for a care team, a good starting point is a short conversation with your staff about what they already use, what worries them and what they would need in order to use it safely. You can also explore our CPD training options as part of planning wider professional development for your team.