Google's free AI training for 650,000 UK and Irish teachers: take it, and keep the thinking in school

On 5 October 2026 Google announced that it is "expanding our free Google AI Educator Series to all 650,000 educators in the UK and Ireland". According to the post on Google's blog, the programme runs teacher-led virtual and in-person sessions with a bite-sized curriculum "built by educators, for educators", and awards micro-credentials and badges to teachers who complete modules. The series first launched to 6 million educators in the United States. Google has also worked with the National Governance Association on an AI Policy Toolkit for school governors and trustees. Sign-up runs through the Google for Education Gemini Roadshow, which EdTech Innovation Hub reported begins in Leeds on 5 October.

The scale is what makes this news: an offer to every teacher in two countries, not a pilot in a few trusts. Google cites Public First analysis that "almost three quarters (74%) of roles in education have the potential to be enhanced by AI", and ties the expansion to the UK government's goal of training 10 million workers in digital skills by 2030. Most teachers who use AI have so far taught themselves. A free national programme from the company that makes one of the tools they use changes who sets the baseline for what a competent AI user in a classroom looks like.

Dan Fitzpatrick, The AI Educator, says any school leader's first response should be to accept the offer. Free, teacher-led training with credentials removes the two barriers that have held staff back: cost and permission. The profession is already ahead of policy on AI, and training that meets teachers where they are beats guidance that arrives after they have worked it out for themselves. His caution is about what training cannot do. A company that builds a tool will teach people to use that tool well. It cannot decide what a school wants AI for, which tasks stay with a human, or how a child's learning is protected when the doing gets faster. That is the school's job, and it comes before the training, not after.

The line Fitzpatrick finds most significant is the one about governors. Boards have been asked to approve AI policies they have had no training to scrutinise; a toolkit built with the National Governance Association puts the question in front of the people accountable for the answer. It will be most useful as a set of questions a board asks its leadership team, and least useful if signed off in one meeting. There is an equity point too: the schools that gain most from free training are the ones that already have a plan. A micro-credential is evidence of attendance; AI literacy is being able to judge an output, spot where it is wrong or biased, and know when not to use it.

For school leaders, the practical step is to set the frame before sending staff: write down in plain language what AI is for in your school and what stays human, including feedback, relationships and professional judgement; then book the training and ask each attendee to bring back one thing to stop doing. For governors and trustees, use the toolkit to ask what data is shared, how outputs are checked and what safeguarding looks like in practice, and ask for evidence, not assurances. In any sector, vendor training is the floor, not the strategy. Outsource the doing, not the thinking.

Fitzpatrick works with schools, colleges, governing bodies and government organisations on adopting AI and building AI literacy, and writes regularly on what AI means for learning and work.

Dan Fitzpatrick is available for interview and comment on this story.