AI is already in most nurseries. It is on the phones in staff pockets, inside nursery-management and observation apps, and quite possibly behind last week’s newsletter. Often nobody has agreed how it should be used. This AI early years course gives practitioners and managers clear rules for using AI tools safely, lawfully and honestly, so they save time without putting children’s information at risk.
It is not an anti-AI course. It shows where AI genuinely helps, where it must never go, and how to write it all into a policy your whole team can follow.
Why safe AI use matters in the early years
Most informal AI use starts with good intentions: a practitioner tidying a learning journal entry, a manager drafting a reply to a tricky email, a student writing up an observation. Each may be sending information about identifiable children to servers outside the setting’s control. What is typed into a public tool may be stored and in some cases used to improve the tool, and deleting your side of the chat does not delete the provider’s copy.
The law is clear about who answers for it. Under UK GDPR, children’s personal data merits specific protection, and your setting, as the data controller, is legally responsible for where that data goes, including into AI tools used on staff members’ personal accounts. Child data entered into a public AI tool may be a personal data breach that must be reported internally straight away and, where the legal threshold is met, to the Information Commissioner’s Office (ICO).
What’s in the course
- AI is already in your settingWhat generative AI is, where it already appears in early years work, and why the stakes are higher here.
- Children’s data and the lawWhat counts as personal data, the data controller’s duties, and when a Data Protection Impact Assessment is needed.
- The golden rulesWhat must never go into an AI tool, personal accounts and free tools, anonymisation and safeguarding information.
- Using AI wellNewsletters, plain-English policies, translations, risk assessment wording and activity ideas, with human review every time.
- Accuracy, bias and safeguardingHallucination, stereotyped output, children’s photos, AI image misuse and more convincing phishing.
- Writing your setting’s AI policyThe seven ingredients of a workable policy and what to do when a slip happens.
- Assessment10 questions. Pass mark 80%. Resit as many times as you need.
What you’ll learn
- The golden rule: identifiable child data never goes into a public AI tool
- Why “identifiable” is wider than a name, and how combinations of details such as room, age, sibling and home language can identify a child
- Why safeguarding information stays out of every AI tool, even when you think it is anonymised
- The safe working pattern: generic in, personal data never in, human review out
- Why observations, assessments and progress summaries are not AI tasks
- How to check AI output for invented facts and gender stereotypes before it goes out
- Why children’s photos never go into AI editing or filter apps, and how to verify unusual requests for money or data by phone
- The seven things your AI policy must cover, from an approved tools list to a review cycle
Who this course is for
- Practitioners, students and volunteers who use, or might use, AI tools for work
- Managers and data protection leads deciding which tools to approve
- Designated Safeguarding Leads who need to understand the emerging AI risks
- Anyone writing newsletters, letters or policies for parents
How it works
- OnlineStudy on any device and pick up where you left off.
- 6 short lessonsAbout 45 minutes in total.
- 10-question assessmentPass mark 80%. Resit as many times as you need.
- CertificateIssued when you pass. NFAQ accredited.
Training your team
An AI policy only protects children if every new starter, student and volunteer learns the rules before they touch a keyboard, so this course works best taken by the whole team. Choose the number of learners at checkout, with group rates from 3 learners, or get it as part of the Training Portal for £5 per learner per month. It pairs well with GDPR for Staff and Online Safety in Early Years.
Frequently asked questions
Can staff use AI to write observations or progress reports?
No. Observations, assessments and progress summaries are professional judgements about identifiable children, and the knowledge of the child lives with the key person. AI is best kept for generic tasks such as newsletters, letters and activity ideas.
Is it all right if I use my own personal AI account?
No. Work data belongs in channels the setting controls. Using a personal account to process information about children is still processing the setting’s data, through a channel the setting has no contract with and no control over.
Can we use AI to translate letters for families?
Yes, for everyday, generic letters with no child details. Where the stakes are high, such as anything medical or legally important, have a native speaker check the translation before it goes out.
What should we do if a child’s details go into an AI tool?
Treat it as a potential personal data breach. Report it internally the same day so the manager and data protection lead can assess it, record and contain it, and report it to the ICO where the legal threshold is met.
What happens if I don’t pass the assessment?
You can resit it as many times as you need until you pass. Your certificate is issued as soon as you reach the 80% pass mark.
Do we need a written AI policy?
The course recommends one. It turns the rules into standard practice, gives staff clear permission, and shows parents that AI use in your setting is considered and controlled. The final lesson walks you through drafting it.
