"In terms of its impacts on people and planet, AI is as physical as any other industry." Thoughtful piece by @imogen-parker.bsky.social reflecting on last week's Global Fund for a New Economy convening in Nairobi: www.globalgovernmentforum.com/lessons-from...

We at @instituteforgovernment.org.uk have coordinated a letter in @thetimes.com today in which numerous individuals and organisations call on the government to withdraw its mistaken rules which are having a serious chilling effect on civil servants speaking in public…

If you don't pay for the product... www.theguardian.com/technology/2...

Government departments and local authorities are looking to accelerate AI use in the public sector. Our new policy briefing presents six findings from six years of Ada’s research on public attitudes towards the use of data and AI in public services. www.adalovelaceinstitute.org/policy-brief...

Brilliant work from my @adalovelaceinst.bsky.social colleagues @imogen-parker.bsky.social and @lauracarter.bsky.social on what the UK public expect from AI deployment in the public sector www.adalovelaceinstitute.org/policy-brief...

Rapidly deploying AI has become one of the government’s top priorities. Read our new blog from @elliotmjones.bsky.social and @imogen-parker.bsky.social proposing four targeted recommendations to strengthen the foundations for AI in the public sector ⬇️ www.adalovelaceinstitute.org/blog/four-re...

If Ada's public services research landed on the desk of those leading digital transformation within government – alongside an extensive to-do list and a myriad of competing pressures – where should they start? @adalovelaceinst.bsky.social www.adalovelaceinstitute.org/blog/four-re...

I failed to post this last week, but it’s one of the best op-eds you’ll read all year. Two politicians with *wildly* different orientations but a common thread of belief in the nature of citizenship and nationality. (They’re also 100% right obvs.) on.ft.com/3GS1rPl

Highly sensitive information, relating to people in moments of vulnerability, has been hacked. The advice warning people of scams is important but only part of it. It's not hard to imagine harms arising if this type of information circulates or is weaponised. www.bbc.co.uk/news/article...

Why public legitimacy isn't just a 'nice to have'. My new column for Global Government Forum on public sector AI - and risks of moving out of step from the public. www.globalgovernmentforum.com/why-public-l...

Interested to see how this pans out...

PAC have published their Use of AI in Government report. Lots of valuable insight, acknowledging potential for transformation while warning Gov "faces significant challenges if it to grasp these opportunities, & we remain concerned at the scale of the task." committees.parliament.uk/work/8580/us...

Want to know how the public feel about AI? We've got you. Here's the next wave of Ada and Turing's UK public attitudes survey published today 📢 attitudestoai.uk

Job alert!! Could you be Ada's new Associate Director in Emerging Tech & Industry Practice, leading a fab team and a highly impactful research programme? It's a pretty amazing gig, @agstrait.bsky.social has built something incredible (no pressure). Get in touch if you have Qs!

“There is this terrible misconception that disabled people just take, but actually by taking from us, you prevent us from being able to give – we want to be able to participate in life equally the same as anyone else, and that includes going to work.” Disability campaigner Tanni Gray-Thompson

Really like this list of questions to consider in the context of public sector AI. And lots of resonance with findings from Ada's research synthesis. A big +1 to sorting out the building blocks.

I'm delighted to say our highly secure and productive OpenSAFELY platform has received a further £17 million in funding from Wellcome, including for a new mental health data project. More here, and clips from Today on Radio4 this morning too! www.bennett.ox.ac.uk/blog/2025/02...

📢 New report out (and offering serious competition for best Ada title). Today we publish Learn Fast and Build Things. It's a synthesis of the 6 years and 30+ publications on public sector data and AI. www.adalovelaceinstitute.org/policy-brief...

X has become an extraordinary case study in irresponsible innovation. The lesson is that carelessness does not just make a technology less nice, it also makes it less useful.

I know everyone was hoping for a quiet Friday, but... UK Gov has announced a step back from AI safety, by rebranding AISI as the AI 'Security' Institute. Really thoughtful commentary from colleague @mbirtwistle.bsky.social here & short 🧵below on key changes www.ukauthority.com/articles/ai-...

Looking forward to the (only marginally less glamorous) UK leg of the Paris AI Summit today. Speaking with a great line up on AI in public services - come say hi if you're around. aifringe.org @adalovelaceinst.bsky.social

The Paris AI Action Summit is almost upon us, but will it actually move the needle towards AI that works for people and society? In our latest blog post we share some of our key questions for the Summit ⬇️ www.adalovelaceinstitute.org/blog/a-piece...

Two pieces of essential reading from Ada colleagues in advance of the AI Summit First, the excellently titled 'Delegation Nation' A primer for policymakers on Advanced AI Assistants by @juulsm.bsky.social @harryfarmer.bsky.social www.adalovelaceinstitute.org/policy-brief...

In the run-up to the French Summit, we've released a briefing on advanced AI assistants and why they should be front and center in the discussions around safety, governance and regulation. See below for details ⬇️

In Paris listening to the fifth talk in a row on global AI risks to humanity, and not one mention yet of runaway AI energy demand pushing us over the climate brink

EU ready to use 'bazooka' trade tool against Big Tech in retaliation against Trump www.ft.com/content/7303...

What are we really talking about when we talk about AI? Latest piece for the Global Government Forum 👇 www.globalgovernmentforum.com/what-are-we-...

We are doing a exciting new inquiry looking at 14-24 year olds' experiences of growing up with technology. We are looking for a partner to support a major piece of peer research as part of the programme. Please share www.adalovelaceinstitute.org/our-work/res...

🤔 The 🇩🇰 government asked us: how can democratic control over tech giants be strengthened? We - the government's expert group on tech giants - outline 7️⃣ principles and a set of guidelines for tackling economic, democratic, and security vulnerabilities:

"The regulation and governance of AI in EdTech has not kept pace with the evolution of the products, leaving pupils and schools exposed to potentially risky technologies being deployed." Hugely important work on AI and edtech in schools from @adalovelaceinst.bsky.social and @nuffieldfoundation.org

I am delighted our education and AI landscape review ‘A learning curve?’ has been published. 🎉 @adalovelaceinst.bsky.social @nuffieldfoundation.org I hope this work will generate further research and evidence to support schools, teachers, pupils and policy makers going forwards.

It was a pleasure to be apart of reviewing this report, and the findings could not come at a more important moment. As we head to the AI Action Summit, global governments face a choice - continue to let the evidence of AI's risks and harms pile up unaddressed, or take action to protect people.

You don't have to see this as a bubble to note the astonishing fragility of expectations in the AI industry www.bbc.co.uk/news/article...

Great to see this out in the open. Dropping AI pilots isn't a bad thing. It underscores the importance of testing new tech before rolling them out. Particularly in welfare, where the risks of amplifying inequalities and causing real world harm are significant. But... (1/5) tinyurl.com/6tn2r24x

this is a dystopia these are the things that happen in dystopias if you'd put into an SF novel 20 years ago that people helped others avoid brutal police raids by holding up handmade signs in online video while cheerily talking about cats... that would be a dystopian novel

Interested in the UK govt’s recent moves on AI? Have questions about how we should respond to the harms, risks and problems these systems pose? Come read this briefing by Demos, the Ada Lovelace Institute and Connected by Data

Bookmarking another AI case study which highlights real-world harms arising from a rush to roll out. AI can do incredible things -which can be socially beneficial. We're much more likely to see positive outcomes if we have effective regulation to balance profit incentives. tinyurl.com/2rythyeu

Uhm, sorry, what? Starmer thinks AI is going to DOUBLE UK productivity in less than 5 years?

Does he think you can just rock up to the jobcentre, say you're depressed and get a load of benefits? He knows that's not the case, so he's purely scapegoating

An additional prerogative with AI is that the AI industry beyond the very few makers of the biggest models really want regulation. They want assurance. They need their work to be trustworthy, otherwise they won't have the certainty they need to invest.

The UK AI Opportunity Plan and Gov's response is now live... There's lots to comb through, but a few first thoughts below 🧵(1/8). www.gov.uk/government/p... @adalovelaceinst.bsky.social

right?

Good stuff in the Pat McFadden speech about public sector reform. Test and learn / accepting uncertainty in particular. But 'tour of duty' construct is deeply unhelpful for anyone looking to build a career in government, as I wrote pre-election about my experience: richardpope.org/blog/2024/03...

A window into the blackbox - longer form thoughts on new latest batch of algorithms published by the UK Government medium.com/@imogencathe... @adalovelaceinst.bsky.social and ht @halcyene.bsky.social

Massive kudos to all the civil servants past and present that have pushed and pushed to get to this point. An incredible achievement and milestone.

Pleased to share the latest version of my paper with Arthur Spirling and @lexipalmer.bsky.social on replication using LMs We show: 1. current applications of LMs in political science research *don't* meet basic standards of reproducibility...

Also, they haven't published all responses. We submitted a return and got an email today to say it will be published in the next tranche in mid Jan.

This morning the UK Gov published a new batch of entries of algorithms in use in the public sector (on their Algorithmic Transparency Reporting Standard). www.gov.uk/algorithmic-... Long form response tomorrow coming, but a quick 🧵 @adalovelaceinst.bsky.social

👀Gov just dropped 14 more entries on their algorithmic transparency reporting standard... combing through now. www.gov.uk/algorithmic-...