Chatbot privacy and compliance: GDPR, the AI Act and more

The privacy and AI rules that apply to business chatbots, including transparency, lawful basis, data minimisation, retention, vendors and the EU AI Act's chatbot disclosure duties.

By the UK Technology Hub teamPublished 4 minute read

The short answer

Chatbots process personal data, so data protection laws such as UK and EU GDPR apply: you need a lawful basis, transparency, minimal data collection, security, retention limits and vendor agreements. The EU AI Act's transparency rules, in force from August 2026, require informing people when they interact with an AI system unless it's obvious. Sector rules add extra duties for health, finance and children.

Every chatbot conversation can contain personal data: names, emails, order numbers, health details or opinions. Using AI to process it brings legal obligations. This guide summarises key requirements. It's general information, not legal advice; requirements vary by country and sector.

Data protection principles

Under UK and EU GDPR, and similar laws elsewhere:

  • Lawful basis: identify why you process chat data, such as performing a contract or legitimate interests.
  • Transparency: explain in plain language how chat data is used.
  • Purpose limitation: use data only for stated purposes.
  • Data minimisation: collect only what's needed.
  • Accuracy: correct wrong information.
  • Storage limitation: set retention periods.
  • Security: protect data appropriately.

Transparency and AI disclosure

Tell users they're talking to an AI assistant. The EU AI Act's transparency obligations, applicable from 2 August 2026, require providers to design AI systems that interact with people so users are informed, unless it's obvious from context, and include rules on labelling some AI-generated content. Other jurisdictions have similar disclosure rules in specific contexts.

Vendors and processors

Chatbot platforms and AI model providers usually process data on your behalf:

  • Sign data processing agreements.
  • Check whether data is used for model training, and opt out where needed.
  • Understand where data is stored and international transfer safeguards.
  • Review security certifications.

Sensitive data

Health, financial, biometric and other special category data need extra care. Avoid collecting it in chat unless necessary, and apply stronger safeguards when you do. Our guide to AI chatbots in healthcare covers health data.

Children

Services likely to be used by children face additional rules, such as age-appropriate design requirements and children's privacy laws.

Automated decisions

If a chatbot makes decisions with significant effects on people, such as eligibility or approvals, specific rules on automated decision-making may apply. Keep humans involved in significant decisions.

Individual rights

People can request access to their data, correction or deletion. Make sure chat logs can be found and managed.

Risk assessments

For higher-risk uses, carry out a data protection impact assessment, documenting risks and mitigations.

Practical checklist

  • AI disclosure in the chat interface.
  • Privacy notice covering chat data.
  • Lawful basis documented.
  • Data processing agreements with vendors.
  • Training opt-outs where appropriate.
  • Retention periods set and enforced.
  • Access controls on transcripts.
  • Process for data subject requests.
  • Risk assessment for sensitive uses.
  • Regular reviews as laws change.

Compliance requirements are part of the wider evolution of chatbots; see our guide to AI chatbot trends for 2027.

A worked example

An online optician launched a chatbot to answer questions about glasses, contact lenses and appointments. A privacy review before launch found that customers might share health information, such as eye conditions, in conversations. The company updated its privacy notice, added a short message at the start of the chat explaining how data was used, limited conversation retention, restricted staff access to logs and confirmed its vendor didn't use conversations to train models. It also added an easy way for customers to request deletion of their chat history.

Writing a chatbot privacy notice

  • What data the chatbot collects.
  • Why it's collected and the legal basis.
  • Who it's shared with, including vendors.
  • How long it's kept.
  • How people can exercise their rights.
  • Who to contact with questions.

Sector considerations

Regulated sectors face extra duties. Our guides to AI chatbots for law firms and AI chatbots in healthcare cover confidentiality and sensitive data in those settings.

Reviewing regularly

Privacy compliance isn't a one-off task. Review data flows, retention and vendor terms when you add new features, channels or integrations.

Anonymising analytics

When analysing conversations, remove names, contact details and account numbers wherever possible. Anonymised data still shows trends without exposing individuals.

Children's data

If children might use your chatbot, follow the stricter rules for children's data, use age-appropriate language and limit data collection.

How we help

Our AI chatbot development team builds assistants with disclosure, minimisation and retention designed in, our AI solutions team reviews vendors and data flows and our web app developers build tools for managing transcripts and data requests.

Frequently asked questions

Does GDPR apply to chatbots?

Yes, whenever chatbots process personal data of people in the UK or EU.

Do I have to tell users they're talking to AI?

The EU AI Act requires it in many cases from August 2026, and it's good practice everywhere.

How long should chatbot transcripts be kept?

Only as long as necessary for their purpose. Set and enforce retention periods.

This guide is general information, not legal, tax or financial advice, and details such as prices, features and rules can change. Product and company names mentioned are trademarks of their respective owners and are used only to identify those products; UK Technology Hub is not affiliated with or endorsed by them.

Related guides

AI chatbots in healthcare: safe uses and key rules

How clinics and health businesses can use AI assistants for appointments, admin and information, the limits on clinical advice, and the privacy and regulatory rules involved.

All AI chatbots & assistants guides