DfE announcement: 450,000 disadvantaged pupils could benefit from AI tutoring tools

The Department for Education’s announcement that up to 450,000 disadvantaged pupils could benefit from AI-powered tutoring tools signals an important moment for the education system. It recognises both the scale of educational inequality and the growing potential of artificial intelligence to provide targeted, personalised support for learners who need it most.

At AI in Education, we welcome this ambition. Well-designed AI tools can help pupils revisit foundational knowledge, receive immediate feedback, and access explanations tailored to their needs - particularly where teacher capacity is stretched. For disadvantaged pupils, including those who may lack access to private tutoring or consistent support outside school, this could be genuinely transformative. However, the announcement also raises a critical question: how will schools ensure that AI tutoring is implemented safely, ethically and in ways that genuinely improve outcomes rather than introduce new risks?

Evidence from schools already experimenting with AI tells us that technology alone does not close gaps. Without clear governance, staff confidence, safeguarding arrangements and inclusive design, AI risks being inconsistently used, or worse, reinforcing existing inequalities through bias, poor data use or over-reliance on automation. This is where a structured, education-led approach becomes essential.

The AiEd Certified Framework was developed precisely to support schools navigating moments like this. Rather than focusing on individual tools, the framework helps schools take a whole-institution view of AI adoption, spanning AI literacy, policies and ethics, tools and systems, digital pedagogy, and collaboration with pupils and communities. Crucially, it places equity and safeguarding at the centre of decision-making, not as an afterthought.

For AI tutoring tools aimed at disadvantaged pupils, this means asking the right questions before scaling up: How are pupils supported to use AI critically rather than passively? What safeguards are in place around data, bias and age-appropriate use? How are staff trained to integrate AI tutoring into high-quality teaching rather than bolt it on? And how is pupil voice included in shaping what “good” AI support looks like?

The DfE’s announcement rightly focuses on opportunity. The next phase must focus on implementation quality. If AI tutoring is to benefit disadvantaged pupils at scale, schools need more than pilots and products; they need trusted frameworks, professional learning and shared standards that build confidence across the system. AI has real potential to support equity in education. But it will only do so if we remain clear-eyed: the impact will come not from AI itself, but from how thoughtfully schools are supported to use it.

Read the full announcement from the DfE here: 450,000 disadvantaged pupils could benefit from AI tutoring tools - GOV.UK