This workshop brings together a deliberately mixed group: AI researchers and developers alongside social scientists, policymakers, journalists and civil servants. The idea is to have an open, collaborative discussion about what “hype” actually is, how it operates in different contexts (a benchmark result, a funding pitch, a policy briefing, a news headline), and what would help all of us reason more precisely about what AI systems can and can’t currently do.

The second half of the day shifts from open discussion to consultation: we want to draw on your expertise to understand how you currently evaluate claims in your own work, where existing tools or norms fall short, and what a genuinely useful hype literacy framework would need to look like to actually get used. Your input will directly shape the frameworks and priorities that come out of the day. 

The day will include:

• An open discussion on what hype is, how it operates across research, industry, media, and policy, and why it matters for evaluating claims under uncertainty

• An interactive session examining real-world AI claims together, grounded in participants' own technical and domain expertise

• A card-based exercise mapping how different stakeholders (including developers and researchers themselves) are affected by AI narratives

• A closing session where we ask participants directly what Hype Studies should prioritise next

Places at this workshop are by invitation only. If you wish to contribute, please email genai@ucl.ac.uk with any expressions of interest.