Data Policy Digest
Hello, and welcome to our 27th Data Policy Digest, bringing you all the latest data and AI policy developments.
The world has changed somewhat since our last edition just over a month ago.
Yes, the Data (Use and Access) Bill has started its parliamentary journey proper.
And there was some big electoral event in the US, global ramifications, end of the world, blah blah blah. More on those stories, and much more, below.
I’ve tried using AI to help compile this edition for the first time (formatting hyperlinks rather than anything more taxing), with some mixed results - it saved some time, but it skipped some links and renamed others. I’m sure it’s just a coincidence it watered down some of the headlines that were most critical of AI…. (Coincidentally, Rachel Coldicutt has just written about how people are using genAI at work.)
Speaking of playing around with the newsletter… we’d love to get your feedback on the Digest. What do you like? What don’t you like? Now’s your chance to tell us via this form - all thoughts very gratefully received.
And… if you’re free next Monday 9 December, we have some festive drinks for participants and friends of the Data and AI Civil Society Network, partner organisations and other friends.
If there’s something we’ve missed, something you’re up to that you’d like us to include next time or you have any thoughts on how useful the Digest is or could be, please get in touch via gavin@connectedbydata.org. We’re on Bluesky @connectedbydata.bsky.social, and Mastodon @ConnectedByData@social.coop. You can also catch up on previous Digests.
To receive the next edition of the Data Policy Digest direct to your inbox sign up here.
Contents
What everyone else has been up to
Data policy developments
Bills, bills, bills
Data (Use and Access) Bill (DUAB): 19 November was the Second Reading in the Lords for the Data (Use and Access) Bill - the first chance for peers to properly debate the substance of the Bill, a debate you can read in Hansard or watch back. The government has also published supporting docs, though not yet a Keeling schedule, which shows how the Bill will change existing laws - legal consultancy Handley Gill have stepped into that breach.
The next stage is committee, where the Lords will go through it line-by-line: dates for that are 3, 10, 16 and 18 December. As with the predecessor Data Protection and Digital Information Bill, it’s Grand Committee - that means amendments (which are already appearing) will not be voted on and only made if agreed unanimously, otherwise they’ll be withdrawn and no doubt reappear at a future stage - and we have the rough order for what will be debated when.
Before the debate, there were published briefings from various members of the Data and AI Civil Society Network - which now has a working group dedicated to the Bill (scroll down for details) - including us… ORG (who’ve also started a petition about the Bill)… Big Brother Watch… Defend Digital Me… ODI… medConfidential… Careful Industries… 5Rights… our resource page on the old Data Protection and Digital Information Bill may also be of interest…
And elsewhere, from the Lords Library… the ICO… techUK… Chris Pounder… and (on a subject that didn’t really come up last time round) Sex Matters… there are also summaries from some law firms, such as… The UK’s Data (Use and Access) Bill – (not quite) the DPDI Bill by another name? (a load of stuff from Taylor Wessing)… What’s New in the Draft UK Data (Use and Access) Bill? (Hunton Andrews Kurth in the National Law Review)… while UK revives plan to reform data protection rules with an eye on boosting the economy was TechCrunch’s summary… and Exclusive: Charities urge government to keep soft opt-in for charity marketing emails (Third Sector, also covered by UKFundraising and Civil Society)…
In terms of the debate itself… Second reading shows promising signs on Data (Use and Access) Bill (ODI)… Westminster update: peers raise our concerns on data protection (Law Society)… UK’s upcoming digital verification bill becomes part of trans rights debates (Biometric Update)… AI concerns punctuate UK data bill debate (Pinsent Masons)… DUA (Lipa) Second Reading: The AI void (Herbert Smith Freehills)…
On subjects covered by the Bill… the Lords Constitution committee is still less than delighted with some of the powers being given to ministers and lack of scrutiny (the Bingham Centre also take a look at delegated powers in the Bill)… Peter Kyle responded to the Lords European Affairs Committee on their data adequacy concerns… he also spoke to the Express about the National Underground Asset Register (DSIT have published a project update on that)… Ofcom has launched a call for evidence on Researchers’ access to information from regulated online services (closes 17 January), while the European Commission are also exploring the topic as part of the Digital Services Act… and lots from DSIT on digital identity, including Digital identity services: inclusion monitoring report findings and UK digital identity and attributes trust framework: supporting documents…
On subjects people might like to see in the Bill, or tried getting into its predecessor… UK government failing to list use of AI on mandatory register (The Guardian) - they’re apparently (still) coming ‘SOON’ - but does the current hub page give us some clues? (It might not!)… Open address data published by Leeds City Council – Ordnance Survey intervenes to protect Royal Mail IP (Owen Boswarva)… No simple replacement for digital evidence rules, says Post Office Horizon trial judge (Computer Weekly).
AI Bill: As per the last Digest: A narrow frontier-focused Bill, putting the AI Safety Institute on a statutory footing and making voluntary agreements between government and industry mandatory, is expected soon, accompanied by a consultation (which we understand may go wider).
Cybersecurity and Resilience Bill: Coming in 2025. Meanwhile… Major Incident Declared at Merseyside Hospital After Cyberattack (Liverpool Echo)… Merseyside Hospital Declares Major Incident Following Cyberattack (Computing)… Thames Water’s IT ‘Falling Apart’ and Hit by Cyberattacks, Sources Claim (The Guardian)… Cyber Incidents Reported in UK Drinking Water Infrastructure (The Record)… and Russian hacking software used to steal hundreds of MoD log-ins (the I).
Property (Digital Assets etc) Bill: Well here’s a treat for parliamentary nerds (yes I know I need to get out more, thank you): since this is a Law Commission Bill (one brought following recommendations of the statutory, independent Commission), there’s a slightly different procedure. The first proper chance to debate the contents was at a Second Reading Committee (before a perfunctory second reading), and it now goes to a Special Public Bill Committee which will take evidence, which started on 26 November (next hearing: 5 December). The Lords Library and Law Society have briefings.
Other King’s Speech Bills: Product Regulation and Metrology Bill is in Lords Committee stage… the Employment Rights Bill is in Commons committee, as the Business and Trade Select Committee launches an inquiry into it, and the Regulatory Policy Committee rated the government’s impact assessments about it red (eight of 23 were ‘not fit for purpose’).
Online Safety Act: As per, the Online Safety Act Network newsletter (they’ve had two since our last Digest) will have much more, but a few things we’ve spotted include…
New online safety priorities for Ofcom and launch of study into effects of social media on children (DSIT)… Draft Statement of Strategic Priorities for online safety (DSIT)… The OSA and the draft Statement of Strategic Priorities (OSAN)… Now Online Safety Act is law, UK has ‘priorities’ – but still won’t explain ‘spy clause’ (The Register)…
Open letter to UK online service providers regarding Generative AI and chatbots (Ofcom)… How children will be protected from accessing online pornography (Ofcom)… Advisory Committee on Disinformation and Misinformation (Ofcom), and Understanding misinformation: an exploration of UK adults’ behaviour and attitudes (Ofcom)… Online Safety Act 2023: RPC opinion (green-rated) (Regulatory Policy Committee)…
Social media ban for under-16s on table if tech giants fail to protect children (Telegraph)… Social media ban for UK under-16s ‘not on the cards’ for now, says minister (The Guardian)… Should smartphones be banned in schools? (FT)… Lords debate on mobile phones in schools (Hansard)… Australia plans social media ban for under-16s (BBC)… Australia passes social media ban for under 16s (Sky News)… Australian-style social media ban for under-16s ‘a retrograde step’, say UK charities (The Guardian)… How Australia’s tough social media ban compares to laws in other countries (Reuters/The Guardian)… Dozens of states ask Congress to un-doom the Kids Online Safety Act (The Verge)…
‘Sickening’ Molly Russell chatbots found on Character.ai (BBC)… Snapchat most-used app for grooming, says NSPCC (BBC)… Man who made ‘depraved’ child images with AI jailed (BBC)… Rise in AI and ‘nudification’ apps aiding child abuse deepfakes (The Times)… ‘It’s so easy to lie’: A fifth of children use fake age on social media (BBC).
Digital Markets, Competition and Consumers Act: The government has launched a consultation on subscriptions implementation under the Act… there’s a consultation from the CMA on Draft new guidance for road fuel monitoring (closes 5 December)… the CMA has written to the government to ask for approval of its new digital markets competition guidance… and DSIT has published its Plan for monitoring and evaluating the new pro-competition regime for digital markets.
Private Member’s Bills: Josh McAllister’s Protection of Children (Digital Safety and Data Protection) Bill has its Second Reading in March - though Sky News reports that a plan to ban smartphones in schools has been dropped. (There’s also this from the BBC which refers to a possible Bill on banning smartphones but it’s unclear which one.)
Other: Bishops seek talks over ‘left-field’ proposal to abolish the Lords Spiritual (Church Times).
DSIT up and take notice
Budget: Not long after our last Digest went to pixel, the Chancellor rendered it instantly outdated with the first Budget from a Labour Government in 14 years. I tried to summarise what it had to say about data and AI for the ODI; there was reaction from IFOW and techUK; DSIT focused on R&D investment in a press release; Public Technology had more on DSIT’s role ensuring a ‘coherent’ approach to digital in the spending review (coming in June, apparently); there are terms of reference for the Review of Technology Adoption for Growth, Innovation and Productivity announced in the Budget; and here’s the now-concluded consultation on open data for road fuel prices the Budget referenced.
The traditional centre: Let’s stick with the Treasury, since it’s still a relevant novelty that they’re thinking about data, tech and AI… Chancellor Rachel Reeves’ Mansion House speech talked a bit about harnessing digital technology (and listening to business - here’s a reminder of the Civil Society Covenant, chancellor - which NCVO are doing some work on) and working with President-Elect Trump on emerging technologies (Politico suggests a ‘mini-trade deal’ is being discussed in Whitehall, and innovation and tech was also in the PM’s message of congratulations)… it also mentioned digital gilts (blockchain klaxon! Remember blockchain?)… the Treasury’s newly-published areas of research interest include novel data sources, emerging tech including AI, and citizen engagement… the Council for Science and Technology wrote to the Chancellor… and ‘CDDO and Treasury develop AI investment metrics for spending review’, reports Public Technology…
The FT reckoned Defra perm sec Tamara Finkelstein, justice perm sec Antonio Romeo, health perm sec Chris Wormald and former Brexit negotiator Olly Robbins were the shortlist for Cabinet Secretary… and it would appear they were definitely right about Wormald, who has just been appointed to the post. No 10 are also looking for a PPS for the PM…
And the Times kicked off reports that a dashboard for tracking progress on the government missions may be imminent (non-paywalled version also available) as part of a ‘Plan for Change’, with other coverage including… Labour ‘needs to tell a better story’ – and Morgan McSweeney has a plan (Guardian)… Bold vision or flubbed reset? Starmer set to reveal targets for this parliament (ITV News)… Keir Starmer to launch online ‘dashboard’ to let public track policy targets (FT)… Keir Starmer lines up Labour relaunch … but is it mission impossible? (Sunday Times)… Keir Starmer’s new delivery targets are intended to give an electric jolt to Whitehall (The Observer)… Keir Starmer must remember his mission (New Statesman)… and Starmer himself wrote for the Sun on Sunday, The machinery of Government is like turning an oil tanker… but we will get there and deliver change for working people. Thursday seems to be the day when we’ll get more details.
The PM also attended a constituency event, as Google launched its London AI Campus, while his biographer Tom Baldwin wrote about how Unlike Blair, Starmer’s a pragmatist on tech (The Times).
DSIT: There’s an update from DG Emily Middleton on shaping the new digital centre (we get a shout out), and from Martha Lane Fox too. It comes as ‘GDS ops chief Robinson takes on role leading creation of “Future DSIT”’… there are some more external thoughts about the digital centre, including three ideas from Sarah Gold and visualising government as a platform from Ben Welby, while Civil Service World republished some initial thoughts on the move (including from me)… and GDS hits several milestones, specifically on One Login (‘celebrating 50 services’, also written up by Public Technology, and ‘making securing and completing an apprenticeship easier’), Forms (New tool to make it easier and faster for public to access essential government services online) and GOV.UK Chat (Government’s experimental AI chatbot to help people set up small businesses and find support - which also notes the source of a lot of the saving time framing of Labour digital policy; How we’re designing GOV.UK Chat; and from the Mirror, ‘I tried the Government’s new AI chatbot - it didn’t like some of the questions I asked’).
As for DSIT ministers… Feryal Clark has been talking about how tech can save citizens time… Maggie Jones spoke at the WTO AI conference… Peter Kyle spoke to The Times about AI safety and much else, including the need for ‘humility’ and ‘statecraft’ when dealing with big tech (which former MEP Marietje Schaake found a ‘baffling misunderstanding of the role of a democratically elected and accountable leader’)… he also spoke to The House magazine about his brief… and outside DSIT, new investment minister, Poppy Gustafsson (also a member of the digital centre advisory panel), was introduced to the Lords… Andrew Western, transformation minister at DWP, opened Government Transformation Week, while Stephen Morgan, early education minister at DfE, covered AI and online safety in a chat with The House.
While we’re on people and processes… the DSIT board has been refreshed… ‘CDDO seeks senior leader to help “develop different approaches for funding digital work” across government’… and DSIT has a new parliamentary private secretary (ministers’ eyes and ears with the parliamentary party) in Matthew Patrick. Other tech-interested MPs have been appointed to other departments, though Lucy Rigby’s stint did not last long - she’s the new Solicitor General following the post-Louise Haigh reshuffle.
In other DSIT-related news… James O’Malley wonders about the National Data Library, Ellie Ashman writes about the lessons from GOV.UK Registers, and Wellcome and the ESRC launch a UK National Data Library: Technical White Paper Challenge… Public Technology reported on the creation of the Office for Digital Identities and Attributes, and how ‘One in four DSIT staffers now based outside London’… the FT had a story, ‘UK visa process for AI experts should be streamlined, says government adviser’ (also summarised by The Register), which previews the Clifford report on AI Opportunities (Politico reckons we might see that this week, while the FT have a piece, Matt Clifford: the UK tech adviser whose influence has drawn sceptics)…
And… Ofcom has appointed Lord (Richard) Allan to its board and made Tamara Ingram deputy chair… RTAU has published the Model for Responsible Innovation, ‘a practical tool created to help teams across the public sector and beyond to innovate responsibly with data and AI’… an update to the Responsible AI Toolkit… Finland: UK Science and Innovation Network summary… New National Quantum Laboratory to Open Up Access to Quantum Computing, Unleashing a Revolution in AI, Energy, Healthcare and More… Repository of Privacy-Enhancing Technologies (PETs) Use Cases… Privacy-Enhancing Technologies Cost Benefit Awareness Tool… The 4th Republic of Korea-UK Cyber Dialogue Held in London… Digital Veteran Card to Make Accessing Support Even Easier for Ex-Service Personnel… UK and Republic of Korea Mixed Committee Meeting on Science and Technology: Joint Statement… Geospatial AI for Land Use by the Alan Turing Institute.
And finally… a shout out to the House of Commons Library, whose list of social media bots includes Bluesky and Mastodon accounts flagging whenever DSIT has answered a written parliamentary question.
Other departments: The ‘Get Britain Working’ white paper (see also: press release) includes AI tools to support job coaching, including one called DWP ASK (which I don’t think is mentioned anywhere else on GOV.UK)… ‘Digital with a human touch’ – DWP sheds light on £200m plan to modernise State Pension and nine other benefits (Public Technology)… Mapping the Future: Geospatial’s role in Mission-Driven Government (Government Transformation).
AI: The summit of AI developers in San Francisco which the UK Government announced back in September has happened - I can’t yet see a summary on GOV.UK, but the European Commission and US Department of Commerce have some details. The AI Safety Institute has also recapped its first year (note it has a separate website from the main GOV.UK one and so won’t show up in search there).
From DSIT… Assuring a Responsible Future for AI, a new report which ‘surveys the state of the UK AI assurance market and sets out how DSIT will drive its future growth’ (DSIT)… consultation on DSIT’s new AI Management Essentials tool (closes 29 January)… Ensuring trust in AI to unlock £6.5 billion over next decade, as UK signs AI safety agreement with Singapore… AI innovators to fuel UK’s clean energy future as new competition launches (‘the second round of the Manchester Prize will focus on delivering new, clean energy systems’)…
Defence… MOD to investigate potential for AI to improve productivity (Civil Service World)… Putin is harnessing AI against us, minister tells NATO (The Times)… UK and its allies must stay one step ahead in new AI arms race (GOV.UK)… Government launches Laboratory for AI Security (Civil Service World)… Government seeks to replicate attacks designed to use AI in developing chemical weapons or election interference (Public Technology)…
Elsewhere in government… How generative AI is accelerating outcomes in DBT (Digital Trade Blog)… AI tool could influence Home Office immigration decisions, critics say (The Guardian)… Government project to test human resilience to AI via deceptive games (Public Technology)… AI transparency: DRCF perspectives (DRCF)… Exclusive: Government trials AI coding tools (Public Technology)… and Scottish MSPs call for AI laws before it’s too late (Public Technology).
Stats: The importance of separation: Ed Humpherson addresses questions raised by the Lievesley Review (Office for Statistics Regulation)… Driving forward improvements to health and social care statistics (ONS Blog)… Health and social care statistical outputs (GOV.UK)… It’s time for a new ONS website platform (ONS Digital, Data and Technology).
About government: Gordon Guthrie’s big independent report for the Scottish Government on ‘Foundations of the Digital State’ has been published (with a launch event and other bits and pieces from Gordon)… and while we’re in Scotland (from October but I may have missed it at the time), there’s a new Digital Capability Advisory Group…
Elsewhere… The productivity paradox redux: From Whitehall to the town hall, technology continues to frustrate (New Statesman)… both the Social Market Foundation, with AI Use in UK Public Services (which will apparently save CENTURIES of time, CENTURIES - Think Digital have a write-up of the launch and CSW of the report), and Demos, with Tech that Liberates: A New Vision for Embedding AI in Public Service Reform (watch the launch, or read a live commentary from Anna Dent), published reports on AI in public services last week… A Future Vision for AI: Public Perspectives on the Role of AI in Public Services (University of Cambridge)… Government Digital Transformation Hampered by Data Sharing Challenges, Research Reveals (Government Transformation)… The Price of Everything: A Plan for the Office for Value for Money (Reform)… Why the NHS Should Have Bloomberg Terminals (Reform)… and the IfG’s new Public Bodies Tracker includes the National Data Library and Regulatory Innovation Office.
Everything else: Home Office Policing Performance Standards Unit Announced (Civil Service World)… Police Cloud Project Raises Data Protection Concerns Despite Legal Reforms (Computer Weekly)… Home Office Creates ‘High-Risk and Complex’ Major Project to Transform Police Intel Database (Public Technology)… New Report Calls for a Single National Police Agency to Better Support Frontline Policing (Police Foundation)… Many unable to access eVisas to prove right to be in UK, Home Office admits (The Guardian)… AI could help us predict prison violence, says justice secretary (The Times)… Smile! UK cops spend tens of millions on live facial recognition tech (The Register)…
ORG Urges ICO to Revise Public Sector Enforcement Approach (Computer Weekly)… ICO Tells Local Government to Show Empathy and Action on Data Protection (Public Technology)… ICO Annual Reports 1985 to Date (Information Rights and Wrongs, plus an extra one from Owen Boswarva)…
The Sudlow Review (HDR UK - you can also watch the launch - a correspondent highlights page 123 showing data flows to Secure Data Environments, and comments at the launch on linking NHS and HMRC data)… NHS technophobes are blocking reforms, warns GP at Hand founder (Telegraph)… Why are doctors wary of wearables? (BBC)… Northern Uses Fax Machines to Send Messages to Train Crews (The Guardian), though see also Analogue Because of Digital (Ben Holliday)… NAO Disclaimed Opinion on Whole Government Accounts Due to Missing Local Authority Data (Civil Service World)… Denmark AI-Powered Welfare System Fuels Mass Surveillance and Risks Discrimination, Report Finds (Amnesty International)… Swedish authorities urged to discontinue AI welfare system (Computer Weekly)… ‘We need to read this stuff and sometimes its awful’: Scottish FOI commissioner on WhatsApp usage (Civil Service World)… and a behind-the-scenes look at how London government is using city data to serve citizens better on electric vehicles (Theo Blackwell, LOTI).
Party people
Hard Labour: ‘EXCLUSIVE: Privacy crusade peer Tom Watson lands job at US ‘spy tech’ firm which is at the centre of an NHS row over patients’ data’, reports the Mail… and while we’re on Palantir, it comes as Good Law Project say Wes Streeting must rule Palantir out of his NHS technology revolution, and… Tech firm Palantir spoke with MOJ about calculating prisoners’ reoffending risks (The Guardian, plus some thoughts from Michael Veale)…
And the Labour Infrastructure Forum has published some policy asks - you’ll need to sign up for the doc, but they include ‘Driving digital growth’ (connectivity, the digital divide, etc) and ‘Making Britain a leader in data centres’ in ‘the midst of an AI revolution’.
Tory story: The US wasn’t the only place this month where a long-running election came to a close with a controversial right-winger winning - yes, Kemi Badenoch (who has a bit of a tech background and has spoken on online safety in the past) is the new Conservative leader. Anthony Painter has an interesting take on something which may link the new leaders on both sides of the pond - a dislike of the ‘bureaucratic class’. With Andrew Griffith moving to shadow business, we also have a new shadow DSIT secretary - Alan Mak, who has previously written about tech and the fourth industrial revolution (this cartoon is a… highlight?), and more recently about how to make the UK a science and technology superpower. Ben Spencer, Viscount Camrose and Lord Markham complete the shadow DSIT team (Saqib Bhatti has moved to DCMS).
And former chancellor Sajid Javid will guest edit Radio 4’s Today programme over Christmas, with ‘items on the benefits and risks of artificial intelligence (AI) as well as the BBC children’s programme Grange Hill’.
Parly-vous data?
Let’s get down to business and start with, well, parliamentary business… In the Commons over the last month or so… 30 October: in his final Prime Minister’s Questions as opposition leader, Rishi Sunak asked the PM to ‘find his inner tech bro and continue to support emerging British tech businesses and establish our country as the home of AI growth and innovation’… 30 October was also Budget Day (see ‘DSIT up and take notice’)… 6 November: written statement from the Treasury on Silicon Valley Bank UK… 12 November: the Science, Innovation and Technology Committee took oral evidence on the 2024 Budget… there was a Westminster Hall debate brought by Kirith Entwistle (Lab) on Tackling Image-based Abuse… 13 November: Westminster Hall debates on Police use of live facial recognition technology (John Whittingdale, Con - the Commons Library has a briefing) and Rural Broadband (Sarah Dyke, LD)… 13 November: written statement from the Home Office on a Consultation on personal liability on senior executives of online platforms and marketplaces regarding illegal content on weapons… 20 November: DSIT questions… and the Women and Equalities Committee took oral evidence on Tackling non-consensual intimate image abuse… 26 November: Westminster Hall debate on ‘Online Safety: Children and Young People’, tabled by Lola McEvoy (Lab)…
In the Lords over the last month or so… 29 October: first oral evidence session for the Communications and Digital Committee’s inquiry into ‘Scaling up - AI and creative tech’ (further sessions on 19 November, 26 November)… 30 October: question from Baroness Ritchie (Lab) on catfishing on social media… AI was one subject raised in a Public Services Committee hearing on interpreting and translation services in the courts (as it was on 13 November and perhaps 27 November)… 31 October: debate from Viscount Stansgate (Lab) on The contribution of science and technology to the UK economy… 5 November: Peter Kyle and senior officials appeared before the Science and Technology Committee… 5 November: question from Baroness Fox (unaffiliated) on sex and gender collection in future Census questions… 11 November: Lord Clement-Jones (LD) had a question on the Vallance Review into regulating digital tech and the government’s policy approach to intellectual property rights and generative AI… 12 November: the Science and Technology Committee took oral evidence on ARIA, the Advanced Research and Invention Agency, from chair Matt Clifford and CEO Ilan Gur… 18 November: questions on digital and AI literacy as part of a curriculum review (Lord Holmes, Con - the Media and Information Literacy Alliance have called for information literacy too) and Improving the use of evaluation in government policy-making (Lord Kempsell, Con)… 19 November: Data (Use and Access) Bill Second Reading (see ‘Bills, Bills, Bills’)… 21 November: debate from Baroness Stowell (Con) on the Communications and Digital Committee’s report, Large language models and generative AI… 25 November: the Communications and Digital Committee published its report on ‘the future of news’, which includes championing responsible AI and updating media plurality rules to take genAI into account (you may also be interested in a DCMS consultation on updating the media merger regime for online news, and the government response to the DCMS select committee on ‘trusted voices’)… 28 November: debate on Schools: Mobile Phones, moved by Baroness Kidron (crossbench).
Coming up in the Commons… 28 November: DCMS questions (including a fair few questions on supporting the creative industries)… 3 December: the Science, Innovation and Technology quizzes Peter Kyle and perm sec Sarah Munby on the work of DSIT… 4 December: PACAC takes oral evidence from the civil service COO on the work of the Cabinet Office… 5 December: Cabinet Office questions… 6 December: debate tabled by the Lord Archbishop of York on the ‘Importance of social cohesion and strong, supportive community life during periods of change and global uncertainty’ which may well touch on AI… 8 January 2025: DSIT questions… 13 January: Westminster Hall debate on a petition related to children’s social media accounts (the BBC has more)… 17 January: second readings of private members’ bills, including the Social Media (Access to Accounts) Bill (Max Wilkinson, LD)…
And in the Lords… 3 December: first day of committee stage for the Data (Use and Access) Bill… also that day, another oral evidence session on ‘Scaling up - AI and creative tech’… 4 December: more on that inquiry into interpreting/translation services in the courts… 13 December: second reading of a couple of private members’ bills, the Non-Consensual Sexually Explicit Images and Videos (Offences) Bill (Baroness Owen, Con), and the Public Authority Algorithmic and Automated Decision-Making Systems Bill (Lord Clement-Jones, LD).
Regulatory nerds - specifically those concerned with how the ICO might deal with duties or strategies relating to competition and growth, as per the new Data Bill - may also be interested in the Financial Services Regulation Committee (Lords) inquiry into the FCA and PRA’s secondary competitiveness and growth objective.
While we’re on select committees: as trailed last time, Lib Dem MPs Tom Gordon and Martin Wrigley have joined the Science, Innovation and Technology Committee, completing the line-up… PAC has (re)opened an inquiry into the use of AI in government… the Business and Trade committee is looking at the Employment Rights Bill… chair of the Science, Innovation and Technology committee, Chi Onwurah, appeared on The News Agents podcast, subjects including Elon Musk and misinformation (she also wrote for politics.co.uk on summoning Musk and has spoken to the Telegraph about the abuse she’s received for doing so)… the committee also has a call for evidence on Social media, misinformation and harmful algorithms (closes 18 December)…
And while we’re on cross-party groups of MPs - more and more details of All Party Parliamentary Groups (APPGs) are being published, including ones on AI, Digital Communities, Digital Identity, Digital Inclusion, Digital Regulation and Responsibility, Diversity and Inclusion in STEM, FinTech, and the longstanding Internet, Communications and Technology (PICTFOR). The Digital Inclusion APPG has published a third ‘state of the nation’ report, while Tortoise reports on the Russian links of one of the members of an APPG considering fair elections and misinformation.
Parliament goes into Christmas recess from Friday 20 December, returning Monday 6 January.
Other bits and pieces… Use of digital, scientific and other technologies by the police and wider criminal justice system (POST)… Better declarations of interest for Parliamentary Questions (mySociety)… Peer Review, a new interactive tool about the Lords (Tortoise)… in Scotland, MSPs call for AI laws ‘before it’s too late’ (Public Technology)… and in Wales, the Senedd’s Economy, Trade, and Rural Affairs Committee has a new inquiry into AI and the Welsh Economy.
AI got ‘rithm
International: There’s some more on the US under In brief (‘States of play’) below, but also… One Year into Biden’s AI Order: Will a New President Change Course? (Tech Policy Press)… ChatGPT, Meta and Google generative AI should be designated ‘high-risk’ under new laws, bipartisan committee recommends (ABC News)…
Elsewhere, Commission Launches Consultation on AI Act Prohibitions and AI System Definition (European Commission)… CDT Europe’s AI Bulletin: November 2024 (Center for Democracy and Technology)… Safeguarding Freedom of Expression in the AI Era (Tech Policy Press)… AI Governance: Mapping the Road Ahead (Computer Weekly)… AI’s Alarming Trend Toward Illiberalism (Foreign Policy)… The Very Real Constraints on Artificial Intelligence in 2025 (The Economist)… the Brazilian Government has a new AI Lab… and Sharon O’Dea looks at a public study from the Office of Digital Government Western Australia on using Microsoft Copilot.
Big AI: Steve Messer has a useful thread of some pieces on possible bubble collapse (Gary Marcus, CONFIRMED: LLMs have indeed reached a point of diminishing returns; FT, AI groups rush to redesign model testing and create new benchmarks; The Stack, Is an AI winter coming? Diminishing returns and scaling limit fears freeze AGI hopes)… see also The Simple Macroeconomics of AI (Daron Acemoglu, NBER)…
Nvidia just became the world’s largest company amid AI boom (TechCrunch)… Artists say they leaked OpenAI’s Sora video model in protest (The Verge)… OpenAI hits pause on video model Sora after artists leak access in protest (Washington Post)… Elon Musk adds Microsoft to lawsuit against ChatGPT-maker OpenAI (BBC)… Why Public Sector Procurement Needs an AI Rethink (Faculty AI)… Reddit CEO thanks AI for helping the site finally turn a profit (The Register)… Microsoft Provided Gender Detection AI on Accident (404 Media)… OpenAI accidentally erases potential evidence in training data lawsuit (The Verge)… AI could be the key to unlocking a more efficient UK public sector (Google Cloud)… Google’s Shadow Campaigns (Microsoft)…
Regulation, law and governance: New rules? Lessons for AI regulation from the governance of other high-tech sectors (Ada Lovelace Institute)… AI Alignment and Governance in Large Language Models (arXiv)… and a special release of Generative AI for Law from MIT, with pieces on Governing Digital Legal Systems, The Alchemy of Legal Data, and An Introduction to Large Language Models for E-Discovery Professionals… Cognitive Scientist Gary Marcus Says AI Must Be Regulated. He Has a Plan. (Wall Street Journal)…
Media and creative industries: AI-generated slop is quietly conquering the internet. Is it a threat to journalism or a problem that will fix itself? (Reuters Institute)… Neither humans-in-the-loop nor transparency labels will save the news media when it comes to AI (Reuters Institute)… Inside the Booming Industry of AI Chatbot ‘Pimping’ (404 Media)… Writers Condemn Startup’s AI Book Plans (The Guardian)… There’s No Longer Any Doubt That Hollywood Writing Is Powering AI (The Atlantic)…
Public sector: From Black Boxes to Meaningful Transparency: AI in the Public Sector (Global Government Forum)… Doctors Are Optimistic About AI in Healthcare (The Alan Turing Institute)… Judge Rules in Favor of School Against AI Use in Grading Dispute (Gizmodo)… ChatGPT Has No Place in the Classroom (Buttondown)…
Everything else: AI and the Future of Election Campaigns: Why Didn’t AI Happen in 2024’s Elections? (Who Targets Me?)… Inside the AI-Powered Propaganda Machine (The Times)… Artificial Intelligence, Scams, and the Dangers of Propaganda (The New Republic)… AI tools enhance judgment of relevance of research to society (FT, check out the comments)… AI for Qualitative Research at Scale (London School of Economics)… Guarding Wikipedia Against AI Hoaxes (404 Media)… Deciphering Alien Language Through AI Communication Analysis (The Conversation)…
What’s the impact of artificial intelligence on energy demand? (Sustainability by Numbers)… The environmental campaigners fighting against data centres (BBC)… Conservationists turn to AI in battle to save red squirrels (BBC)…
Character.ai and Eating Disorders: How AI Chatbots Impact Mental Health (Futurism)… Brands Avoid “AI” in Marketing to Build Customer Trust (CNN)… Mind the Gap: Foundation Models and the Covert Proliferation of Military Intelligence, Surveillance, and Targeting (arXiv)… Email and the Rise of AI in Day-to-Day Business Practices (404 Media)…
A new golden age of discovery: Seizing the AI for Science opportunity (Google DeepMind)… Is AI Dominance Inevitable? A Technology Ethicist Says No (The Conversation)… Can AI Improve Itself? The Quest for Self-Improving AI (Ars Technica)… Are the robots finally coming? (Financial Times)… AI Isn’t About Imagination; It’s About Outsourcing for Profit (The Guardian)… Between Hype and Hesitancy: How AI Can Make Us Safer Online (Tech Policy Press)… Ethical AI Deployment: A Philosophical Perspective (Philosophy Documentation Center)… Kenyan Perspectives on Shaping the Future of AI (Ada Lovelace Institute)… Mapping global approaches to public compute (also Ada)… State of Compute Access 2024: How to Navigate the New Power Paradox (TBI - who also have new reports on The Impact of AI on the Labour Market, and Revitalising Nuclear: The UK Can Power AI and Lead the Clean-Energy Transition accompanying A New Nuclear Age)… Beyond ‘AI boosterism’ (Progressive Review)… Future of AI Summit (Financial Times)… Stewarding a greater ecology of accountability in the age of AI (Trinity College Dublin’s new Artificial Intelligence Accountability Lab)… How the Rise of New Digital Workers Will Lead to an Unlimited Age (Time)… Automating Summation — On AI and Holding Responsibility in Relationships (Bianca Wylie)…
And… Pokémon Go Data Adding Amplitude to War is Obviously an Issue, Niantic Exec Says (404 Media)… Coca-Cola Holiday Ads Use AI, and the Internet Can’t Stop Complaining (The New York Times)… Deus in Machina: Swiss Church Installs AI-Powered Jesus (The Guardian).
In brief
Here’s looking at EU: The European Parliament has approved the new Commission (more from Politico), including Executive Vice-President for Tech Sovereignty, Security and Democracy, Henna Virkkunen.
If you want to catch up on how her confirmation hearing went, then… Hearing of Executive Vice-President Designate Henna Virkkunen (European Parliament)… European Commissioner Hearings: Henna Virkkunen on Tech Sovereignty, Security, and Democracy (Politico)… EU Technology Security Chief Henna Virkkunen’s Hearing and Elon Musk’s Role (Politico)… EU Commissioner Hearings: Glenn Micallef, Maros Sefcovic, and Henna Virkkunen (Politico)… Europe’s New Tech Commissioner Actually Likes Tech (Euractiv)… Virkkunen Advocates for Investment, Cutting Red Tape in EU Telecoms (Euractiv)… DigitalEurope’s Reaction to the Virkkunen and Sejourne Hearings: An Act is Not the Answer to Everything (DigitalEurope)… and Politico has graded the previous Commission…
And elsewhere… Massive hack-for-hire scandal rocks Italian political elites (Politico)… Apple reportedly facing first-ever EU fine over App Store rules (The Verge).
States of play: You may have noticed there was an election in the US, and that former President Donald Trump emerged victorious.
Everyone is scrabbling to understand what that might mean for… well, everything, including tech, data and AI. We had this on the Republican election platform back in July:
The Republican Party says ‘We will repeal Joe Biden’s dangerous Executive Order that hinders AI Innovation, and imposes Radical Leftwing ideas on the development of this technology. In its place, Republicans support AI Development rooted in Free Speech and Human Flourishing’… this is apparently part of championing innovation, which also includes opposing the Democrats’ ‘unlawful and unAmerican Crypto crackdown’, and opposing a central bank digital currency…*
It may also be worth checking out Project 2025 - the Presidential Transition Project - whose ‘Mandate for Leadership’ has a good few references to AI and tech.
Other reflections include… Scoop: Trump eyes AI czar (Axios)… Trump, Harris, and the US Election: What Does it Mean for NATO and Digital Diplomacy? (Politico)… What Donald Trump’s Win Means For AI (Time)… What Trump’s Presidency Means for AI (Global Government Forum)… Trump plans to dismantle Biden AI safeguards after victory (Ars Technica)… Musk’s Rivals Fear He Will Target Them With His New Power (Wall Street Journal)… How Tech Companies Will Fare Under Trump (Fast Company)… Tech Leaders Under Trump: What’s Changed for Bezos, Zuckerberg, and Pichai (The Verge)… The Coming Merger of Tech and State Power (Tech Policy Press)…
Additionally… Are Fears About Online Misinformation in the US Election Overblown? (Reuters Institute)… The Sensitive Personal Information Behind Gunmakers (ProPublica)… Infrastructure for the Digital Age: Building a Safer, More Resilient Digital Ecosystem (New America)… Exclusive Research: How the Next US Administration Can Lead on Digital Service Delivery (Global Government Forum)…
Musk odour: Elon Musk has been a prominent figure in the Trump campaign. He has been selected, alongside Vivek Ramaswamy, to head a Department of Government Efficiency. Yes, that’s two heads of DOGE - very efficiency, many horror, wow. (Or not quite Cerberus, if you fancy a more elevated reference.)
Plenty of reaction to that and more, including… Musk, Government Cuts, and the Election’s Impact on Tech (Washington Post)… Musk, Ramaswamy, and Doge: Trump’s Newest Plan (Washington Post)… Trump, Elon Musk, Vivek Ramaswamy: A Plan to Make Government More Efficient with Doge (CBS News)… Trump, Musk, Ramaswamy, and the Deep State (NPR)… they’re hiring, too… Shame on the Elon Enablers (Disconnect Blog)… Elon Musk’s X as a Political Weapon (The Atlantic)… How X users can earn thousands from US election misinformation and AI images (BBC)…
The UK continues to attract Musk’s ire… Elon Musk, Brand and the UK’s Tyrannical Police State (Politico)… Elon Musk, UK Government, and the Politics of Stalinism (Politico)… How Elon Musk and Trump’s Friendship Fell Out in the UK (Politico)… Oxford Scientist Resigns Over Musk’s Continuing Fellowship (The Guardian)… Why I Have Resigned from the Royal Society (Dorothy Bishop)… Big Tech in the UK (William Cullerne Bown)… Elon Musk is attacking Britain. Former PMs are building ties (FT)…
Antisocial media: Elon Musk: The Future of His Twitter (Financial Times)… Congress Investigates Left-Wing Think Tank and Musk’s Twitter (The Telegraph)… Why The Guardian is No Longer Posting on X (The Guardian)…
How to Customize Your Bluesky Profile (The Verge)… EU Says Bluesky Violates Information Disclosure Rules (Reuters)… Bluesky’s Rise as a Twitter Replacement (The Atlantic)… Someone Made a Dataset of One Million Bluesky Posts (404 Media - and update)… I’ve spent the past almost-decade studying research uses of public social media data… (Dr. Casey Fiesler)… Down in the Posting Mines (Marie Le Conte)… Bluesky breaching rules around disclosure of information, says EU (FT)… and a reminder if you’re signing up to Bluesky, there’s a Starter Pack and List with various members of our Data and AI Civil Society Network…
The UK government apparently has no plans for departments to join, but they can use TikTok in exceptional circumstances…
TikTok’s Romanian reckoning (Politico - and more)… ‘Why I spent my university fees on Somali TikTok battles’ (BBC)… It’s Important to Talk About Online Abuse (The Guardian)… The Images of Spain’s Floods Weren’t Created by AI (The Guardian)… Instagram Plans to Use AI to Catch Teens Lying About Age (Bloomberg)…
And… LinkedIn: We’re too boring for kids for social media ban (BBC).
Everything else: Rogue’s Gallery: The History of Facial Recognition Technology (Slate)… Moneygram replaces CEO after massive customer data breach (TechCrunch)… Pregnancy Tracking App ‘What to Expect’ Refuses to Fix Issue that Allows Full Account Takeover (404 Media)… Post Office halts development of IT system to replace Horizon (Financial Times)… Sat-nav companies make changes after fatal crash (BBC)…
Sell Chrome to end search monopoly, Google told (BBC)… The 2024 Future Perfect 50 (Vox)… Mapping South Asia’s Digital Landscape (Atlantic Council)… What are banks doing with your financial data? (Financial Times)… Anyone Can Buy Data Tracking US Soldiers and Spies to Nuclear Vaults and Brothels in Germany (Wired).
What we’ve been up to
- You can now read a write up of our Connected Conversation on 17 October, about Generative AI and Worker Rights
- On 13 November, Tim spoke as part of a panel for UNICEF UK’s in-house Digital Week on the Risk and Benefits of AI
- On 20 November, Tim (and Margaret, from our People’s Panel on AI) took part in a workshop on People, Participation and Power, organised by Liverpool Data Co-operative, ESRC Digital Good Network and Elgon Social
- We’ve published a blogpost by Margaret on reforming public services with tech, and from Care Full on how Looking for Care Revealed a Data Desert
- CIGI published a piece from Jeni, on Why We Need Inclusive Data Governance in the Age of AI
- And a couple of weeknotes each from Emily and from Jeni, and one from Tim
And coming up…
- 3 December, with Big Brother Watch and Public Law Project as a prelude to a new group as part of the Data and AI Civil Society Network: How AI is affecting welfare rights - a civil society primer
- 11 December: an online workshop as part of our Data and AI Public Engagement Community of Practice, on what do public engagement practitioners need to succeed?
- 16 December: we’re supporting Gloucestershire Data Day
- And… we’re one of the partners in the Participatory AI Research & Practice Symposium being held alongside the Paris AI Action Summit - you have until 8 December if you’d like to submit a proposal.
What everyone else has been up to
- Involve published the late Simon Burall’s Thoughts on Democracy
- As it happens, Policy Lab blogged about People-Centred and Participatory Policymaking and held an event with Demos on Art, Design and Participation in Public Policy, New Local had a new report about communities shaping local development, Involve published Could Involving Citizens Help Unlock True English Devolution?, Future Governance Forum asked What Does It Mean to Govern in Partnership?, there were some reflections on participation from Tim, and PEDRI have a ‘New look, same commitment to good public engagement in data research’
- Access Info announced the death of their executive director and founder, Helen Darbishire
- Privacy International have been busy, with… From chatbots to adbots: sharing your thoughts with advertisers… their work, including Non-fitted devices in the Home Office’s surveillance arsenal: Investigating the technology behind GPS fingerprint scanners, featuring in ‘AI’ tool could influence Home Office immigration decisions, critics say (The Guardian)… Muddied Waters: Why (Data Protection) Tribunals Must Explain Their Factual Assessments… Digital health, big tech and your privacy… and Action Against Privacy Risks
- mySociety discussed a change in language, with Pro-Democracy Tech: Reframing Civic Tech for the Current Moment
- Why is my air fryer spying on me? Which? reveals the smart devices gathering your data - and where they send it (Which?)
- What good is innovation if it doesn’t work for everyone? (Promising Trouble)
- Ada held an event on The UK’s Industrial Approach to AI and published Now you are speaking my language: why minoritised LLMs matter
- The Royal Society and DeepMind held a forum on AI for Science
- The ODI pondered What Are Data Spaces and What Do They Do? You can also catch up on recent videos of an event with Richard Pope on his book, Platformland, and the policy team looking at the road ahead for data and AI policy this parliament
- ORG published the ICO Alternative Annual Report 2023-24; Data Orchard published State of the Sector: Data Maturity in the Nonprofit Sector 2024; IFOW published Data on our minds: affective computing at work; there’s also Open Movements, Commons Causes (Open Knowledge Foundation); and the latest UK-EU Divergence Tracker: Q2-Q3 2024 (UK in a Changing Europe)
- The AI Faith & Civil Society Commission have a new video series, Faith Bytes
- Rootcause have published Winning the Fight for the Future of Information and (previously) Sword & Shield: AI, democracy and the future of information
- Congratulations to the winners of the OpenUK awards
- Luminate have a new CEO
- The Women in Tech Policy Network has launched
- Understanding Patient Data have kindly published all five parts of my long read from the RSS’s Significance magazine, The Story of the R Number: How an Obscure Epidemiological Figure Took Over Our Lives
- Some proper impact from the National AIDS Trust: Working with regulators to protect your HIV data from breaches, Ripple Effect: The Devastating Impact of Data Breaches (ICO)
- And CILIP are looking for a Senior Marketing, Communications and Public Affairs Manager.
Events
- 3 December: How AI is affecting welfare rights - a civil society primer (Connected by Data, Big Brother Watch, Public Law Project)
- 3 December: Mind the gap: bridging innovation’s supply and demand in the AI era - Professor Neil Lawrence (Bennett Institute)
- 4 December: Digital Ethics Summit 2024 (techUK)
- 4 December: Launch of the Blavatnik Index of Public Administration (Blavatnik School of Government)
- 4 December: Think Data for Government (featuring Jeni and me)
- 4 December: Data Bites #51 (Public Digital) - you can also catch up on #50 (and #1 to #49), and sign up for #52 on Thursday 6 February 2025
- 5 December: TransformGov Talks #2 - catch up on the first event with Jeni, and Andreas from i.AI
- 6 December: Rethinking the Inevitability of AI (University of Virginia)
- 6 December: Citizens Advice Data Insights: Spotlight on Gender, with Jess Phillips MP
- 9 December: the third Turing Lecture, Can we live with AI?, with Shannon Vallor (following Can democracy handle AI? with Jonathan Bright, and Can we trust AI? with Abeba Birhane)
- 9 December: Some festive drinks for participants and friends of the Data and AI Civil Society Network, partner organisations and other friends (Connected by Data)
- 10 December: Data-centric AI webinar #6: Chris Johnson, Chief Scientific Adviser, DSIT (ODI)
- 11 December: Data and AI Public Engagement Community of Practice - third meeting, workshop on what do public engagement practitioners need to succeed? (Connected by Data)
- 11 December: AI Openness Round Table (OpenUK)
- January-March 2025: The Age of Mistrust (British Academy)
Good reads
- The Beginning of the End of Big Tech (Wired)
- Acute Psychosis and the Power of Avatar Therapy (The Guardian)
- Doctor Who: The Mystery Behind the BBC’s Biggest Show (BBC News)
- Leveraging Data to Protect Democracy (Project Syndicate)
- Algorithms Are Coming for Democracy, But It’s Not All Bad (Wired)
- UnitedHealth Denies Mental Health Care Using an Illegal Algorithm (ProPublica)
- The Cult of Microsoft (Ed Zitron)
- Polls Are Out, Listening Is In (Prospect)
- Are policies like medicines? (James Plunkett)
- Dead behind the eyeglasses (Hugh Wallace)
- The Big Idea Is Convenience, Making Our Lives More Difficult (The Guardian)
- The Fantasy of Cozy Tech (The New Yorker)
- Not My Problem: I asked an AI to predict the near future of art and creative work. This is the story it told me. (Noema)
- A tech news podcast to help you sort through the wreckage (System Crash)
- And also on podcasts… How Do You Assure AI in the NHS? (Ethics Untangled)
And finally:
- AI hoax hijacks Dublin Halloween parade (BBC News)… Dublin Halloween Parade Hoax: How AI Fake News Fooled Ireland (CBS News)
- Strava: The exercise app filled with security holes (Le Monde)… Strava’s Problem: Fitness App Can Locate World’s Most Powerful People (The Guardian)
- Death by a Thousand Roundtables (Alex Chalmers and Anastasia Bektimirova)
- Best anecdote from working in NHS data… (Syd E)
- The AI Exorcist (Terence Eden)