Weeknote 3.0
The fun continues
And no, I’m not referring to Brexit (although that, too, is very much ongoing…). We’ve been pressing on with our usual mix of conversations, reading, writing and whatnot. More on all that below.
But speaking of Brexit (you can tell what I was working on before this), we did have the International Trade Secretary, Liam Fox, in for a public conversation on Wednesday which you can watch back here. He apparently upset Jeremy Corbyn by failing to rule out accepting chlorinated chicken from the US in any future trade deal, but impressed our in-house trade nerds with his detailed responses to audience questions. His speech, which focused on the process of setting up the Department for International Trade (DIT) after Mrs May became PM in July 2016, also touched on the department’s work to make better use of data:*
‘Another challenge was the need for DIT to become a credible, data-driven, intelligence-led and more efficient organisation. This meant developing our analytical, statistical and data science capabilities, and rolling out a range of new surveys and data collections to inform the development of trade and investment policy.
We laid the groundwork to build solid relationships with the ONS and HMRC to deliver the necessary trade statistics and pushed forward with the OECD and WTO cutting-edge initiatives developing new measurements of digital trade and Trade in Value Added.’
All very sensible, given the discrepancies that are common even between data sets on trade in goods — let alone trade in services, which is much harder to keep track of. The ONS did a good write-up of their work on this back in October, which is worth a look. Combined with the DIT-ONS Data Science Campus collaboration to automatically analyse free trade agreement consultation responses, there seems to be a lot of data/tech-oriented thinking going on in this part of government atm.
It’s not just Dr Fox who’s been talking about the need for better data to improve our understanding of the world and inform policy making, either. Last week, Chief Economist at the Bank of England, Andy Haldane, gave a fun lecture focused on the potential for using more localised data in massive simulations to improve our understanding of the economy. Well worth a read.
Three things that happened this week
1. That explainer we’ve been talking about on the different ways future tech is being used in government is nearly there. Going through a few final tweaks and edits, but should be online very soon. And we’re still open to comments on the first iteration of our glossary of technical terms. Thanks to everyone that has fed back to us so far, in person or by email (comments in doc also fine!) — expansion is on its way.
2. We’ve been scoping up a piece of work that should offer a more wide-ranging overview of the tech landscape in government, delving into the details of some key technologies (in a way there isn’t space for in a glossary or explainer) and exploring the work and roles of the various bodies involved. Depending on how things go, we might publish it. We might also just make it open online as a resource for anyone who’s new to this stuff, and to give further context to our bigger reports that will follow in the coming months (for a reminder of what those will be, see Weeknote 1.0).
3. Doteveryone published a report summarising their research on the views and experiences of workers in the UK tech sector (with their underlying data available here). Quite a few interesting findings, but two that have been difficult to shake: 59% of those working in AI had experienced decisions they thought could have a negative impact on people and society, and 45% of those in the wider tech sector felt their work was subject to too little regulation (only 14% thought the sector was regulated too much).
People we chatted to
- Hans Kundnani popped over from Chatham House for a chat. He runs their Commission on Democracy and Technology, which is doing some interesting work on the interplay between technological change and broader political developments. You can see and submit your own responses to their core research questions here. We also had a good discussion about think tanks experimenting with new, more open ways of working (with his project, Doteveryone publishing the data underpinning their report and, er, this thing you’re reading now all being examples).
- We also had a chance to catch up with the team from the Association of Chief Executives and the Public Chair’s Forum to discuss what’s going on with future technology in the public body space. Understandably, given the broad range of roles and responsibilities that public bodies hold, there is a lot of potential for technology to change them!
- Lewis popped along to the launch event of the National Crime Agency’s National Security Assessment of Serious and Organised Crime on Tuesday (lots on how technology gives criminals new ways of committing and hiding crime, at the same time as creating opportunities for those trying to tackle criminal activity) and Equifax’s Spark Conference (lots on Open Banking and public trust).
What we’re reading and thinking about
- Gavin has been training up the new IfG recruits in the art of beautiful charts (big shout out to Aron, too!), and is getting ready for the next edition of Data Bites (last edition here) — keep your eyes peeled for an invite going out next week for 4th June, especially if you like geospatial data! He also went to his first meeting as a member of the research commissioning board for an ESRC project on making more of government’s administrative data. Data sharing (and whether data sharing is the right phrase, it almost certainly isn’t) is going to be everywhere over the next few months…
- Marcus has been learning about how South Korea is using artificial intelligence to manage export controls and how VR could transform government services. He’s also been working his way through this OECD paper on automation. He appreciates how it provides a critical response to the idea that you can map the complex way that computerisation is/will change work into literal numbers of “jobs at risk”.
- Lewis has been filling his free time catching up with talks from ICLR 2019, which took place in New Orleans last week. Highlight is still this Emily Shuckburgh talk on using machine learning to improve our understanding of the climate and better predict future risks.
What’s coming up next week?
- Things are looking a little quieter. The Lords Science and Technology Committee is taking evidence on science research funding on Tuesday morning, and we’re presenting our thoughts and plans for this project to our razor-sharp colleagues on Friday (wish us luck!).
Any last thoughts?
- * In the last few minutes of the event, in response to an audience question, Dr Fox also touched on his desire to make the UK’s data policy more liberal once outside the EU, and make ‘the ability to move more data’ one of the UK’s asks in future trade negotiations with certain countries. What this means for data protection, alignment with EU countries we might want to continue to share data with, and so on, was not immediately clear. But a good reminder of the wider geopolitical and trade context to discussions about data and technology…
- As ever, hit us up with any thoughts or feedback! Email (digital@instituteforgovernment.org.uk) or Twitter or commenting on this or whatever works best for you — it all works for us.
Till next time!
