FCA authorised · FRN 724952 0117 325 0027Quote & buy →
Aipedia: Social workers
Using a public AI chatbot on your own phone or laptop for casework risks a confidentiality breach and may leave you without your employer's protection. BASW warns that liability increases when AI is used on a personal device without permission.
Part of: AI in social work
In short
Public AI chatbots are free, quick and on every phone, so it is tempting to use them to tidy a case note, draft a letter or summarise a report. Social Work England's research found some practitioners had put personal information into public AI tools. That can breach standard 2.6 on confidentiality and 5.6 on lawful and ethical use of technology, and may be a personal data breach. BASW's guidance says accountability rests with the social worker and liability increases if AI is used on a personal device without employer permission, which goes directly to whether the employer will stand behind you.
SWE's research found that AI use in social work is uneven and employer policies vary. Where an employer has no approved tool, or the approved tool is slow, practitioners under pressure may turn to a public chatbot on their own device. The research recorded that some practitioners had put personal information into public AI tools.
Common examples include pasting a case note to improve the wording, asking for a letter to a parent, summarising a long report, or translating a message. Each can involve personal and special category data about children or adults at risk.
The main duties that apply are:
BASW's guidance, as reported by Community Care, says accountability rests with the social worker and that liability increases if AI is used on a personal device without employer permission. It advises against entering sensitive information without explicit, informed consent. See the professional standards.
Public chatbots may store what you type, use it to improve their models, or process it outside the UK, depending on the service and settings. Personal data about a service user entered into such a tool without a lawful basis is likely to be an unauthorised disclosure. Social care records hold special category data, so the consequences can be serious for the person and for the employer, which may need to assess whether to report the breach.
A personal device adds further risks: no employer security controls, shared family use, backups to personal cloud accounts, and loss or theft.
Employed social workers usually rely on their employer's arrangements if a claim arises from their work. Those arrangements generally assume you were acting within your role and following employer policies. If you used an unapproved tool on your own device against policy, the employer may treat it as a disciplinary matter, and its position on supporting you may be less certain. BASW's point that liability increases without employer permission reflects this.
Most employed social workers do not hold personal PI, so there may be nothing else to fall back on. For the wider picture on data breach liability, see data breach liability for professional firms.
Illustrative scenario: a social worker pastes a section of a child protection report into a public chatbot on their phone to tighten the wording, including the child's name and school. A colleague notices and raises it. The employer treats it as a data breach, assesses whether to notify the regulator, and starts a disciplinary process. Because the incident concerns the social worker's own conduct, it is also referred to Social Work England.
Independent social workers are the data controller or processor for their own work and carry the breach risk themselves. A PI policy may respond to some claims arising from a confidentiality breach, while cyber insurance usually deals with breach response costs; check how your policies fit together. See GDPR and data breach cover within PI, cyber insurance and AI risks and independent social workers: does PI cover AI?.
Put these controls in place:
Related: AI translation with families and Social Work England and BASW on AI.
Apex arranges PI for independent social workers. A broker can explain how your policy treats data breaches involving AI tools; see the AI in social work hub.
Removing names reduces risk but may not remove it, because combinations of details can identify a person. Check your employer's policy first. BASW warns that liability increases if AI is used on a personal device without employer permission.
It depends on the employer's policy and the circumstances. Employer arrangements usually assume you acted within your role and followed policy. Using an unapproved tool against policy may make support less certain and could lead to disciplinary action.
Tell your manager or data protection officer promptly so the employer can assess the breach and any reporting duty. Delaying makes things worse. Being open is consistent with standard 2.1 and is likely to be viewed more favourably than concealment.
Tell us about your firm and the work you do. A broker will review how your current PI treats AI-assisted work, or arrange new cover. No fee for a quote. Or call 0117 325 0027.
Start a PI 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.
