I started this blog during my PhD studies to read civic tech research critically. It gave me a structure for analysing what was being claimed, how it was measured, and whether the methods held up. That helped me hone my research approach. I hope it also helped donors, implementation teams and policy staff make sense of bleeding edge evidence and why it matters.
I’m picking this up again in the same spirit, but with a wider lens.
Coming posts will look at the intersection of technology and society more broadly, with particular attention to AI and digital sovereignty.
The general approach will be the same: trying to cull evidence from hype and hand-wringing, I’ll keep reviewing emerging research and evaluations, flagging when high-level claims aren’t supported by the work that was actually done.
I’ll also take a closer look at how we use AI in research and policy work. Researchers, policy teams and practitioners now use AI to read, code, summarise and score, often without much thought about what that does to the work. I want to look at how different approaches to these tools shape their outputs, and where more standardised and rigorous methods can improve both efficiency and quality.
There’s so much slop out there. It shouldn’t be hard to do better.
If you have thoughts on this, or want me to take something on (a mini lit review, or an assessment of something specific), get in touch. I promise to hallucinate less and get to the point faster than whatever agent(ic team) you’ve already asked.
