We discussed creating workshops, the challenges in using policies and guidelines and applying them to our individual schools.

• We need to design assessments that call for critical and creative thinking - we are seeing more students memorising assessments (when asked to do it hand-written) that were assisted by AI at home.

• The assessment aspect is in our control... this is something we can work on. What may be more concerning is how children are using chat bots. How do we continue to develop in this space? How much of a role should school's play? Children are leaving a digital footprint - student information is being shared with AI (often by the student's themselves). This needs a whole community approach.

• Should we be resourcing specialised subjects like 'Digital Literacy' t ensure students understand the ethical issues

• Special provisions - eg. voice prompts. Whilst it may be an ethical issue that we should respond to disabilities, are we also entrenching dependency... eg. not building up the ability to write in these situations. AI exacerbates these issues as it takes a role in 'doing' work.

Digital dementia and cognitive offloading

Digital social media chat bots

Parents and community

It is great to prompt it for another way of teaching, but you also need to know what the source of truth is (ie Grok has been changed MANY times to now spout "alternative facts". If you can't trust the source, you can't trust the information.

No longer can you google a question and have a trusted, safe website offering advice, but now it is all generative answers, taken from all sources of the internet.

It is much harder to protect students from misinformation, both at school and at home, when AI still gives bad advice. Recently, a student was struggling with self-harm thoughts, and the advice provided by AI was almost the exact opposite of what should have been provided.

Lots more development needs to be done, to keep the information accurate and safe. It isn't there as yet, and that's a concern.

• CoPilot licencing is impossible to work out! Documentation says Agents can be consumed without licences, but this is not what a couple of schools were reporting.

How can we use AI to improve our use of student data in a way that reduces teacher workload?

• Parent teacher interviews: we need to have one source of summarised student data ready for PTIs

• NAPLAN data - would be good to get a summarised snap shot of stduents usable for teachers

• Bringing everything together in one place is important - and that would reduce the workload of looking in lots of places

• We want the AI to be able to interpret the data and give us options of next steps

• There's so much adata but no teacher has the time to look at teh data - and often lack the skills

• AI can synthesis the data and digest it and present it in the right format.

• AI needs to be able to build itself... it needs to be able to set itself up and create systems for us to use

• AI needs to write the integration to join our data together - as an agent. Connecting Ed Perfect, all the places students do work, learning behaviours etc. Then it needs to interpret and summarise.

• AI needs to help us to know our students better... go through their work and give us some key talking points

• AI needs to build a structure that all ed tech companies can align to

Key question: how are students performing versus how are they expected to perform? AI needs to be able to give teachers this answer, with some suggestions as to next step actions.

AI also needs to give a synthesised interpretation of qualitative data.