Experimental

Creating tools for scaling generated content

A flow diagram. The plan is then checked by a rule engine with 21 criteria and a model acting as judge with 19 criteria. This gives a score for each plan, such as 54 out of 56. One loop lets you change the prompt and regenerate a single plan. The other reads all the failures for a rubric and suggests one change to the prompt.

The GOV.UK AI team shares their experimental work on curating content for users to create personalised experiences. They share how they've been using tools to measure the quality of scaled generated content.

What it’s like to run a gov design jam

A small of group of people working together during the Gov Jam, in focus: two young women of different ethnicities, one holding a LEGO figure, coloured sticky notes are behind them on the wall

End of October, public servants from multiple government organisations co-organised Gov Jams in Blackpool, Leeds and London. The hackathon-like events were part of a global event series that took place in 26 cities internationally. In this blog post, 3 organisers share how it worked, what it was like, and how to get involved.