How AI Works in Intelligent Evidence
Our AI reads your uploaded documents and suggests where the content might provide evidence against your framework's grade descriptors, but it never makes the final call. That's always down to you.
What the AI does
It scans uploaded documents and suggests possible evidence, matching it to relevant grade descriptors, categories, and strength levels. It's designed to spot the positives, helping you see where your existing documents already demonstrate good practice, so you're reviewing suggestions rather than starting from a blank page.
What the AI doesn't do
It doesn't approve anything on your behalf. Every suggestion sits in Upload & Review until a person accepts, edits, or rejects it.
It doesn't judge whether your school actually meets a standard, only whether your evidence has been worded in a way that suggests it might. That's why we always recommend checking suggestions against the real data, particularly for things like performance figures or comparisons against a threshold.
It doesn't handle safeguarding evidence. We've deliberately kept this out of the platform, both because of the risk of identifiable personal information, and because safeguarding is a compliance judgement your school and Ofsted are best placed to make directly.
It doesn't work from live or constantly changing sources like websites or shared drives, only fixed documents you choose to upload, so what's been reviewed and approved can't quietly change afterwards.
Why human review still matters
Because the AI is designed to find the positives, that's genuinely useful, but it also means every suggestion needs a second look before it goes into a report representing your school to Ofsted, governors, or your MAT/LA. This is your school's story, told on your terms. We make the suggestions, you make the decisions.