Trust, technology and the future of assessment: Reflections from IAEA 2026

Attending the International Association for Educational Assessment (IAEA) annual conference in Toronto was a great opportunity to hear how assessment organisations and researchers around the world are thinking about the future of assessment.

The conference theme was ‘Trust, transparency and technology in educational assessment’. While technology and AI featured heavily, what struck me was how quickly these conversations became much broader discussions about trust, fairness, validity, responsibility and human expertise.

Trust in a changing assessment landscape

For me, trust was one of the strongest themes running through the conference. With technology and AI developing so quickly, many of the discussions came back to the same fundamental question: how do we innovate without losing trust in assessment?

The keynote speakers approached this from different perspectives. Dr Eunice Eunhee Jang encouraged us to think beyond the traditional assessment triangle towards a wider ‘assessment ecology’, recognising the connections between learners, educators, AI systems and institutions. Dr Kadriye Ercikan explored the pressures AI creates for validity, fairness and comparability, while Dr Mary Richardson brought the focus back to learners and the relationship between assessment, identity and trust.

A session that particularly resonated with me was ‘Between automation and responsibility: The conditions for trust in AI-based assessment’. It connected closely with the work we are doing in AQA’s AI for Assessment Innovation team and the wider question of how we take advantage of new technology without losing sight of responsibility, human judgement and confidence in the assessment process.

More than a technology challenge

AI was obviously a major theme at the conference, but I didn’t come away feeling that the conversation was simply about AI. The programme also explored culturally responsive assessment, transparency in scoring and measurement, student and educator voice, data sharing, privacy and security and the balance between accountability and support.

What I found particularly interesting was how interconnected these issues are. Discussions about technology quickly became discussions about quality, fairness, responsibility and human expertise. It was also reassuring to see that many of the questions we are exploring at AQA are shared across the international assessment community.

One of my main reflections from the conference is that we need to think about innovation as an assessment challenge, not just a technology challenge.

A lot our research is focused on understanding where AI can genuinely add value and where human expertise remains important. As technological capability increases, some of the most important questions are not simply about what AI can do, but what role we want it to play in assessment. Where does human judgement add value? How do we establish validity and quality? Who remains responsible for decisions? And what evidence do we need before changing established assessment processes?

Our role in shaping the future of assessment

Dr Thierry Rocher’s keynote also offered an interesting way of thinking about our work. He described three roles for people working in assessment: the metrologist, the messenger and the guide. This captures something important about our role at AQA. We are responsible not only for measuring accurately but also for communicating what assessment evidence means and helping to shape how assessment evolves.

I came away from the conference with so many fresh ideas for what we could explore. A key challenge is to make sure our research continues to connect technological innovation with our wider expertise in validity, fairness, quality and human judgement.