Connected by data

Why we’re winding up in 2027

Mon, 30 Mar 2026 00:00:00 +0000

When I set up Connected by Data, I also stated my intention that this organisation would have a maximum lifespan of 3-5 years. We are now at the end of our fourth year, which means we have at maximum one more year to go.

It’s always interesting to see people’s reactions when they learn this. The leaders of other civil society organisations often immediately say, without me providing any explanation, “OMG, you can do that? Amazing!” People we work with often say, “That’s a shame, I wish you wouldn’t.”; other people are simply curious and come to see the rationale once I explain it.

In this post I want to explain why I set up Connected by Data this way, what I’ve learned about the trade-offs that entails, and what we’re thinking about doing next.

Working in a changing context

The first explanation I’ll give is the one I usually give to funders.

We work at the intersection of technology and society, and that intersection is always changing. When I was first putting together plans and proposals for Connected by Data in late 2021, we were having important debates about the use of personal (and non-personal) data for collective and public benefit. These came off the back of the experience of Covid, where using sensitive data about things like health, travel and finances was crucial for navigating the pandemic.

The ability for those affected by the uses of that data to have a joint say in the decisions about how it was used – collective data governance – was the thing I most wanted to explore. I also thought the evidence and argument for these approaches was already in existence (from think tanks and researchers) and what was needed was a catalytic campaign to bridge the gap between what they said we should be doing, and actual policy and practice.

Then ChatGPT was released in November 2022, after we’d been operating for about six months. Suddenly the much more pressing conversations were about the data used to train generative AI and the impacts on creative and other workers. Writing now, in 2026, everyone is having these kinds of conversations, alongside ones about digital sovereignty in a geopolitically unstable world, the systemic impacts of the democratisation of generative AI, and so on and so on. There’s very little hard evidence about anything and big differences in thinking about the correct courses of action.

Mission-driven organisations always have to walk a line between focus and adaptability. Too much focus, and your work becomes irrelevant; too much adaptability, and you get swept along with the changing context, never able to target your efforts long enough to make an impact. This is a particular challenge for those of us at the intersection of technology and society because both are changing so rapidly.

So, my rationale for making Connected by Data a fixed term organisation was to give a way to navigate this tension. I knew the problem that I’d identified in 2021 would not be the problem that existed in 2026. I wanted to give myself and the organisation the focus of the 2021 problem and be able to properly re-evaluate, including deciding on a completely different focus or approach, in 2026. Having a fixed term organisation is a way of doing that.

Part of the reason this was the best approach for us is that we’re very small. Larger and longer-term organisations can handle having a fairly consistent core capability, surrounded by projects or campaigns that run around particular issues that come and go over time. Even so, my observation is that it’s very hard for even projects or campaigns to end, and ending is important to free up attention and resources to focus properly on the next thing.

Creating the right incentives

Mission-driven organisations have two big drivers: their mission, and money. Yes, we all want to have the impact in the world that we strive for, but the reality is that we and the people who work with us need money to live. That can mean we take on projects skewed towards what funders think are the right things rather than what we believe are the right things. It can mean we are in competition with other civil society groups for those resources, so are incentivised to downplay their work. We tend to all swarm into the same areas (because those are the ones we can get funding for) but in uncoordinated ways.

But we as civil society desperately need to be coordinated, because it’s only together that we have any hope of being an effective countervailing force against the interests of industry and authoritarian governments.

Being fixed term does not at all obviate the need for funding, but it’s a way to manage those incentives that arise from money rather than mission, and instead insert incentives that mean our work is more collegiate, collaborative and mission aligned.

To work with others

Being fixed term means we can better live the maxim “one can do a great deal of good in this world if one does not care who gets the credit for it”, because we don’t particularly care about Connected by Data brand recognition. It means we are motivated to work with partners who are then able to continue work in an area as we know we’ll be moving on to other things. We actively try to put the structures in place that mean we can leave without negative impact. We make things open because why would we want to hold onto them?

There’s a flip side to this, however. Working in partnership depends on good networks, trust and social capital, which build over time. We’ve had an advantage of already bringing some of these into the organisation, but I can see that a longer lasting organisation would have different advantages with collaborative working.

It’s also not straightforward to catalyse collaborative working amongst a set of other actors. The Data and AI Civil Society Network now has almost 150 members on its Signal group, and meets monthly. I believe it’s proven useful in enabling basic coordination across UK civil society in this space. But we are struggling in our attempts at encouraging others to step forward in that network, to shape, lead, or just basically administer it. It’s hard to tell whether that’s because routine meetings aren’t that useful, or whether there’s an assumption that we will continue to do that lifting. Actually shutting down will at least remove that expectation and hopefully enable others to step forward if we have inadvertently been holding them back.

To not be led by funding

I found it easier to raise mission-aligned money early on with Connected by Data, because I knew I could scale the organisation to the money raised rather than the other way around. I didn’t have to go after funders who weren’t interested in supporting what we wanted to do. That doesn’t mean funding has been plain sailing or straightforward: the Shuttleworth Foundation, from whom I got our foundational fellowship funding, closed after two years, when usually they would have funded for three. Funders like Omidyar Network, Luminate, JRRT and the Mohn Westlake Foundation have all supported us and then moved on to other things. But being fixed term has alleviated a lot of fundraising pressure, and meant we could spend time and effort on pursuing the mission rather than finding money.

This again isn’t a straightforward positive, though. Bootstrapping funding is difficult for new organisations for two reasons. First, some funders want to see existing proof of impact, and proof of good governance such as audited accounts, before funding. Young organisations obviously don’t have either of these; I was relying on my track record at ODI, with funders who knew that work.

Second, many funders understandably put limits on the percentage of an organisation’s income they can fund. (I believe the rationale is that they don’t want to leave the organisation unviable if they don’t continue that funding.) When you are starting, your income is low, and low percentages of low numbers are also low. The only way we managed to bootstrap into having a £400k-ish turnover within this short period was because of the big, unconstrained Shuttleworth Fellowship funding, and funders who were prepared to stretch their rules.

Being fixed term is also difficult from a funding perspective at the end of an organisation. Many funders want to invest in something that has long term impact, and if you’re not going to be around you’re not going to be there to continue to build on the work you did with their funding. We have had rejections from funding that we would have seen through to the end of the funding period, because we won’t be around after it, even when we had other targeted routes for continuation of that work.

Ending well

A third set of reasons for why I set up Connected by Data as a fixed term organisation was because I wanted to end it well. So many civil society organisations come to a close in an extremely stressful rush, and because of that they leave their colleagues, partners and collaborators picking up the pieces. They go through months of uncertainty, hoping for a miraculous new injection of cash, and don’t have time or resources for archiving, reflection or mourning.

Here, I am very much influenced by Stewarding Loss. As they put it:

In a world addicted to growth and permanence, the courage to close—to hospice, to dismantle, to reconfigure —is a radical act of love. And to do this with intention and care, when so much is unravelling, is one of the most generative things we can do - freeing up energy, resources, and imagination to make room for what’s truly needed now.

I have led organisations through radical restructuring. I have worked with organisations as they are closing. I have many CSO CEO friends who are constantly anxious, because they care for their teams and just don’t know how long they can continue to hold them together.

Ending isn’t bad, but uncertainty about ending is extremely destructive. It haunts you, hangs over you, gnaws at you. You can’t concentrate on having an impact with your work when you have to deal with that dread. This is why civil society leaders I meet understand my rationale for setting up a fixed term organisation so easily.

A fixed term at least provides certainty. It gives people time to find other roles (we’ve deliberately built into how we operate that we’re all part time, to provide space for taking on other roles that can ease that transition). It means we have time to pass on or archive our work thoughtfully, so that it’s not lost. It means we can celebrate and mourn together, so our closure is something we can move on from.

It is sad. I love this work and this team. I would continue working forever in this way if I could. But part of what makes this possible now is not having to struggle to sustain it. And I am absolutely convinced that more will grow from our decomposition than would if we continued.

Understanding myself

Finally, of course, there are the selfish reasons. I know that I personally thrive in newness, variety and change. I am much much happier operating in a space where I’m creating something, experimenting and drawing out patterns and lessons, than I am in doing the same thing day after day.

Organisations change as they grow. The startup phase only lasts so long. (There is a fairly well-known phenomenon that mission-driven organisations tend to experience a kind of crisis – and often new leadership – around seven years in, as they hit that transition from new to established.) Even if none of the above reasons applied, and we weren’t going to close, I personally would not continue to lead Connected by Data past this stage because I know it wouldn’t suit me and I wouldn’t be good at it.

I know I’m also far better at caring for people – supporting them to thrive and grow unbound by organisational needs – than caring for organisations. For me, organisations are ephemeral. They are tools we use to organise ourselves temporarily, but our relationships with each other last far longer and are far more important. (Others will see organisations as permanent and our involvement with them as temporary stewardship; I understand and appreciate this perspective too.)

Ultimately, setting up Connected by Data as a fixed term organisation is a choice I made because I knew it would work for me, and I was in the incredibly privileged position of being able to do so. It would not be the right choice for other people, who are more satisfied by growing an organisation and seeing it thrive as an entity. I admire those people; I just know it is not me.

What’s next?

Good question.

We’ll be publishing our annual strategic roadmap soon. Within it there are the definite wrap-up activities, such as archiving and handover, that you’d expect. But we’ve also been encouraged by our Board to explore whether there’s anything that could make use of the organisational entity we’ve created, to avoid the set-up costs of a new organisation. I’m framing these as “regenerations” in the Doctor Who sense. These frankly feel a bit like cheating; I’m wrestling with feeling like I’m avoiding the clean close I’ve argued and aimed for. Nevertheless, over the next six months we will be exploring:

whether we could or should continue hosting collective initiatives like PAIRS, that need an administrative home, through some kind of fiscal hosting arrangement
whether we can usefully provide a loose umbrella organisation for people working as independent consultants or practitioners around collective AI and data governance, public participation, and so on
whether there’s any appetite from philanthropic funders for a programme to support grassroots groups who are affected by and campaigning around data, digital and AI

After these six months of exploration, we will have a clearer answer to “What’s next?”

Thanks for reading this far, and for being along for this ride. We still have a year – a whole fifth of our lifespan and plenty of time to do more amazing things – so this is not yet goodbye!

But I would love you to get in touch if:

you want to make use of our team’s expertise over the next year – it’s your last chance with us in this configuration!
you want to pick up and continue any of our work – we can work together on a proper handover
you’re interested in informing or being part of any of the regenerations I’ve mentioned above – let’s discuss!
you want to bring any of us as individuals into your organisation – we can start that transition before the year is out
you’re interested in the pros and cons, challenges and opportunities of running a fixed term civil society organisation – always happy to chat

Tim Weeknotes

Fri, 20 Mar 2026 00:00:00 +0000

Friday afternoon. An hours train journey. And nothing pressing on my ToDo list. Perhaps an opportunity to finally finish some weeknotes!

What I’ve been up to

Looking back on PAIRS

PAIRS - the Participatory AI Research and Practice Symposium in in Feb was really fantastic. As Octavia Field Reid put it, our symposium day in Delhi was real soul food - and it’s continued to be so encouraging to see all the LinkedIn posts and blog posts continuing to come out reflecting both on the substance of the content, and the space created for people. I feel so grateful to be able to work in a context that calls out the best of our shared capacity for combining rigorous work with a nurturing and caring approach. It has meant a lot to see how much the caring vibe of PAIRS has resonated with people.

On Tuesday morning I published the latest PAIRS Newletter including our Chairs Report and links to recordings and highlights.
I’ve been updating the website with recordings from the online day (all edited by Emily), and editing and uploading recordings from our India sessions this week.

For the India recordings we’ve got a mix of formats, including backup Zoom screencast recordings (the camera crew only arrived mid-morning) and webcam footage, and professionally recorded speaker videos, but without the slides in shot. I’ve been having to use a mix of methods to make sure videos sync slides and videos. The best case turns out to be when our professional recording and zoom recordings overlap (just two timed videos to sync!) and so I really wish I’d left the zoom/screencast recording going all day, rather than stopping after the camera operators arrived! Learning for next time.

This week (and over the last few weeks) I’ve had some lovely calls as well with people who either attended PAIRS, or have since learned about it, and are interested in how they can get involved in the community. Particular thanks to Franziska Heuschkel who put together some fantastic user-journey feedback on what it was like to engage with PAIRS.

I was particularly happy to see this piece in the Indian Press, by PAIRS presenter Ashwin Upreti, and Member of Parliament, Sasmit Patra, titled Society bears AI risks. It must have a say in AI governance, as ripples of our discussions on participatory AI in India continue to surface. Over last weekend it was also great to see submissions go into the AI for Good building on collaboration between PAIRS presenters.

Looking ahead to the Global Dialogue

A lot of my week has been taken up looking ahead to the UN Global Dialogue on AI Governance in Geneva in July, and working on the design of a kick-off meeting for the Citizens Track on AI project.

Thanks to Susan Oman sharing her speaking slot, I was able to talk about this work briefly in the Informal Stakeholder Consultation on the Global Dialogue on Artificial Intelligence Governance on Wednesday - and it was great to hear a number of the contributors talking about the importance of citizen voice.

I’ve also had some helpful responses to my blog post on Expanding the Visibility of Public Voice on AI - What if? and looking forward to picking up conversations on what an Expo presence for citizen voice at AI for Good could look like with a few people next week.

CoCoDa - Research Data Access

Today I was down in London for a workshop hosted by The Open Data Institute and the Data Transfer Initiative focussed on the CoCoDa project that is exploring how researchers can study online platforms “Moving beyond computational methods without (legal) teeth, and legal solutions that are not implementable.”. It was refreshing (for much of the day at least) to get back to a focus on data governance, rather than AI, and to explore one of the core challenges that gave rise to Connected by Data: the necessity, and inadequacy, of individual data rights as a means of governing data for the public good.

There were lots of resonances with the issues we explored with MyData Global, Datasphere Initiative and Aapti Institute in In this together: Combining individual and collective strategies to confront data power - and the point that collective governance is an important complement to individual consent in identifying legitimate uses of data landed well. However, the event also raised some critical questions for me on the distinction between research that seeks data from private platforms:

For ‘social signals’ - to inform general social science research;
For safety / platform impacts research - that uses data to support governance of the platforms the data comes from.

These may need quite different strategies, and techo-legal approaches - yet the debate appeared to move between them quite fluidly.

Green Tech & Talk

I was formally elected as co-convenor of the Science and Technology Policy Working Group for Green Party of England and Wales at our brief AGM on Tuesday. There is lots of energy and expertise pouring into the Green Party right now - and it’s raising some fascinating (and good to have!) challenges around managing passionate debate on the impacts of tech and AI in particular.

Over the coming weeks I’m hoping to get some headspace to think about how we might apply principles of public deliberation and dialogue to some of our member-led policy discussions.

What I’ve been reading

In Monday’s workshop on a Citizens Track on AI Governance, one of the opening inputs will be presented by Iñaki Goñi, so it was serendipitous that my LinkedIn feed served up a link to Iñaki’s latest paper: Design principles of dialogue about science and technology and the design frictions they reveal: Towards a demo-technical analysis. It’s a fantastic timely account of the design tensions in public dialogue on technology, and well worth a read of the whole paper, although Iñaki also offers a wonderfully actionable synthesis of the main analysis in ten questions for the dialogue designer to grapple with (along with a call for more transparency and reflexivity in answering them…):

Locality: How can dialogue methods be adapted to mobilise local priorities while accommodating stakeholder tensions and power differences?

Intra-community difference: How can design engage with participants before dialogue to inform decisions based on their identities and framings of the issue?

Multi-level embedding: How can dialogue bridge different governance levels and create connections between local, regional, and national decision-making?

Methodological standardisation: How can standardised approaches ensure consistency while allowing for flexibility and contextual adaptation?

Personal connection: How can dialogue create a personal connection between participants and the issues at hand, using, for instance, first-hand experiences or artistic engagement?

Open futures: How can dialogue open up imaginative possibilities for socio-technical futures rather than reinforcing existing assumptions?

Meta-dialogical: How can dialogue make itself an object of discussion, encouraging participants to reflect on and critique its own design?

Powerful citizens: How can dialogue help participants recognise and leverage their political power beyond the dialogical space?

Opened-up technoscience framing: How can dialogue surface and explore ideological conflicts in technoscience rather than closing down contestation?

Structured expertise: How can dialogue reshape who counts as an expert and redefine how expertise is engaged with and challenged?

From Goñi (2026) Design principles of dialogue about science and technology and the design frictions they reveal: Towards a demo-technical analysis. Design Studies, Volume 104, 2026.

What I need to pay attention to

Next week is my last week working before two weeks of leave, so I’m keen to keep plans moving for the Citizens Track on AI - as well as to follow up on fundraising activities.

Signatures have now closed on the updated Open Letter on Critical AI Literacy and so I’ve agreed to send that off early next week.

I also need to follow up planning for our webinar on Student Voice on AI in Education on April 22nd, and preparation for a panel on Workers Rights, Green Politics and Technology at the 2026 Hawkwood May Day Festival

Lastly, I’ve got some practical work to do for my local non-CbD projects with Common Good Gloucestershire, and supporting thinking about data infrastructure with Fund Gloucestershire (which I learnt last week was named by the master of ‘Giving Things Names that Describe What They Do’ of ‘TheyWorkForYou’ and ‘FixMyStreet’ fame…).

Emily Weeknotes

Tue, 17 Mar 2026 00:00:00 +0000

What I’ve been doing

I’m writing these weeknotes on a train home from London - I’m both tired but also it seems like a good time to reflect on the last couple of weeks. Also it is worth noting that it is a Saturday. Yes, I have just worked … on a Saturday! What makes me chuckle most about this is that for the ten years I worked in public libraries working on Saturdays (and indeed Sundays) was a fortnightly experience. But gracious my Friday night felt odd.

The reason for working today was a half day student workshop getting their views on monitoring of AI chatbots. I was on safeguarding, parent hosting and logistics. All seemed to go well and people seemed happy (holding my breath a little that I don’t get any calls regarding travel home for them being disrupted). As always we tried to show as much care to the attendees (and their parents enabling their attendance). We had a room for parents to hang out in (if they didn’t want to go for a stroll around London) with snacks - and of course the same for the students.

In addition to sorting the logistics for today I’ve also been spending some time doing video clipping from the hours of recordings of the online PAIRS event and published the first one.

The biggest tick on the ToDo list though was our conference last week. We had some great feedback from attendees and I loved hearing the speakers too. It all went smoothly from a logistics perspective (some comments about the lunch could’ve been better but for a free to attend conference I was comfortable with the quality) so I’m pleased it’s been successfully delivered.

What I need to take care of

Well firstly, finishing these weeknotes. I started them on Saturday and am now finishing them (I hope) on a train back into London today (Tuesday) for a stakeholder version of Tuesday’s workshop. I’ll be notetaking during this session - and bringing the post its. I’m bracing for a lot of post it transcribing on the train home and tomorrow (the turnaround on the write up that Jeni needs to do is very tight).

I also need to finish the PAIRS video clipping. We’re getting close now. There’s a couple where one speaker doesn’t consent so I need to wrangle those a little and work out best how to still platform the rest of the speakers but lots are now on YouTube and by the end of next week at the latest I really hope everything that can be published will be.

Sometime soon I need to start the annual report process. At this point that’s just pulling together everything we’ve done as bullet points, links and photos, into one document. I suspect that isn’t going to start until April though with all the other priorities at the moment.

It’s the end of another quarter this month - and of the financial year - and therefore of the current strategic roadmap. We’ll have a team retrospective accordingly (though we don’t all have coalescing available time in our diaries for almost three weeks now) and then the new roadmap will be published and we’ll be starting our final year of Connected by Data. End of the year also means there’ll be some bits to do for our accountant and a March newsletter to our subscribers is due too.

I can … almost … feel the breathing space …

What I’ve been inspired or challenged or moved by

I haven’t got any links I can share with this update - but our conference was motivating to me. The speakers (including their breadth of perspectives) and the participants who came (stayed) and engaged with all the content. It was a great day.

What I’ve been reading

I’m now reading ‘Sisters of Sword and Shadow’ by Laura Bates. She is the Founder of the Everyday Sexism Project and through her fiction writing challenges misogynistic history - rewriting it with a feminist lens. If you like Arthurian legend fiction (and aren’t offended by strong women) I recommend it.

Expanding the Visibility of Public Voice on AI - What if?

Fri, 06 Mar 2026 00:00:00 +0000

At 1:45 IST on Friday 20th February, at the British Pavilion in Hall 14 of the Bharat Mandapam complex in New Delhi (hosting the India AI Impact Summit), something exceptional happened. For a few minutes, voices of students from a school in Luton played out of speakers, and delegates nearby had a chance to hear their nuanced views about how generative AI in education affects them, and how they want to see it governed in future. This moment formed part of a panel curated by the University of Sheffield sharing work on public voices in AI, and showing videos and photos of everyday citizens sharing their views on AI.

This presence of direct testimony from those affected by AI should not be exceptional (and I’m sure there were at least a few other examples across booths and panels at the Impact Summit). Yet, I don’t think it would be an exaggeration to say that it made up less than 1% of the content that delegates might have encountered. Amidst the glowing blue screens, expert panels, tech demos and glitzy stands, opportunities to engage with evidence from the kinds of informed public engagement that should be shaping our future alongside AI were near non-existent.

That’s a problem.

A few days earlier, at the online and in-person days of the Participatory AI Research and Practice Symposium we heard over 90 presentations on in-depth work across the world to give voice to affected communities in shaping the development and governance of Artificial Intelligence. Rich findings, and critically evaluated methods, that show what different communities want from AI, and how they are encountering and adapting to the impacts of AI currently on offer. The problem is not a shortage of insight from the public - although undoubtedly there is much space for the participatory AI field to grow, and participatory practices to be embedded into business as usual. The issue is that this insight remains far too easy to simply miss or ignore.

So what can we do about it?

Notwithstanding the concern that some actors in the AI space are incentivised to wilfully dismiss public views, I’m left feeling that it is, at least partially, on us to create more opportunities for the AI community to encounter public voices and views. Which, from conversations with many collaborators around the Summit, suggests three ‘What ifs?’

What if we brought together insights from existing public engagement in one place?

We’ve advocated in the past for the new Independent International Scientific Panel on AI to have a focus on evidence from public engagement. It’s likely the first reports will not extend to this in depth, so how could we collectively bring together evidence and insights from existing public engagement on AI?

The Public Voices in AI programme has done methodological groundwork on understanding what evidence tells us, and the open call for papers for PAIRS has shown we can surface many examples of practice across the world. Can we build on learning from projects like Participedia and Assemblis to build a global case book of public views on AI?

What if the expo area at the next big AI conference hosted a showcase of public attitudes to AI?

Although it was great to be able to spend 30 minutes sharing citizen voices at the British Council organised UK Pavillion right at the entrance to the Impact Summit Expo, what if we could bring together groups who have been running public engagement on AI across the world to take over a large expo booth on our own: providing a hub for exploring public attitudes to AI? Imagine a large expo stand sharing videos, posters, literature and chances for conversation with public engagement projects from across the globe.

Would this help delegates to approach other stands and exhibits with impact on affected publics at the top-of-mind? Could this help place citizen voices more central to conference discussions?

What if there was a citizen hub at an AI conference that facilitated direct dialogue between members of the public, and government, industry and NGO delegates?

There are 100s of members of the public who have gained confidence in talking about AI through past involvement in public deliberation, and who would be well placed to share authentic ideas, desires and insights from their community. An expo space or parallel hub could host one-to-one and group discussions, perhaps drawing on the model of the Kavli Center Hopes & Fears Lab .

I think back to my experience many years back helping run a youth-zone at the Internet Governance Forum in Sharm-el-Sheikh with Diplo Foundation, where our booth included both a front-of-house space with comfortable sofas for discussions between young people and delegates, and a ‘behind the scenes’ space, which acted a safe-base where young people perhaps daunted by the dynamic of an international multi-stakeholder conference could debrief, regroup and to head out to engage with conference sessions. I also think about the potential of virtual presence to connect conversations in the conference venue, to publics across the globe.

A shared ‘citizens hub’ (inside or alongside conference/summit venues) could give confidence to NGOs working on public participation to bring public participants along, and could also provide a space for self-registered attendees to engage their citizen identities in how they approach the wider conference.

Even though PAIRS helped us to find many of the people working on public voice around the India AI Impact Summit, I have the sense there is still the need for a meeting point to further help projects and practitioners thinking about public voice to discover each other, connect, and have confidence in their collective voice.

Making it happen?

Although these ideas could be applied to any number of AI events, in exploring them I’ve been particularly thinking about the UN Global Dialogue on AI Governance taking place parallel to the ITU’s AI for Good Summit in Geneva with its large expo.

However, it is fast approaching - so I want to quickly test out whether anyone else can see potential in these ideas.

At Connected by Data we work to catalyse projects that give communities a powerful voice in the governance of data and AI, but we’re a tiny team: so would only be able to work on something like this with partners. I’m already excited to be working with Susan Oman at University of Sheffield on exploring the case book idea, and hoping we might bring PAIRS partners on board, but it will take more of us to really put public voice front-and-centre.

It might be a long-shot to pull something together for AI for Good in 2026, but we can also look ahead to 2027 too.

So - this is a call out:

Would you be interested in collaborating on a presence for public voice at the UN Global Dialogue in Geneva in July?
Do you have ‘Yes and’ ideas that could build on this? Or past experience that might guide us to the most effective approach?
Do you have evidence from public engagement in AI that could be included in a casebook or expo showcase?
Could you help with funding, fundraising and/or logistics?

Drop me a line, or ideally fill in this quick form here before 23rd March and we’ll be ready to follow up!

Emily Weeknotes

Thu, 26 Feb 2026 00:00:00 +0000

What I’ve been doing

One down, one to go. Except the one down (PAIRS which was a couple of weeks ago and a success by all accounts) is not completely done as I still have all the videos to clip / consent and upload (around 15 hours of recordings) and we’ve now added a rapid-turnaround project which includes two in person events (smaller scale but one includes travel logistics for all attendees from across the country). So maybe more accurate to say…two third of one down and now three to go!

PAIRS - The feedback from both the online and in person sessions has been good. There are things we’d do differently next time (there always are) but the people that attended seem happy with the quality and quantity of input and even a couple of technical hitches (at both events) didn’t distract from the motivating vibe from those that engaged. I’ve started doing the video clipping but need to find more time for that. It’s a tricky thing to create time for as it really needs to be a sole focus job and with lots of other urgent things to be doing it keeps getting pushed. I’ve got a long train journey next week though where I’ll be able to do some even if I can’t then upload them until later. The other ‘lag’ action for PAIRS is processing all the travel bursary reimbursements. This is a time sucking exercise (though very worthwhile to do) particularly as the claims are being paid to individuals across the world - and frankly banks / finance systems don’t make those transactions easy. I understand that they have to protect against fraud and laundering and stuff, but heck - it shouldn’t be so hard to give someone the money they’re owed.

Our conference - now with over 80 people registered and just over a week to go we’re getting deep into logistics. Sorting travel support for those that need it (including booking trains for those that can’t be reimbursed) and wrangling other support costs (and trying to draw a fair and compassionate line on what people want / what we can afford with a balance against how important it is for that person to be in the room). As I’ve previously recorded - the “cost” of public participation sits massively in the “enabling” bucket. If you want the public to be engaging then you have to cover their costs…and with no-one underwriting/subsiding those (i.e. an employer which is the case for most professional attendees) then the costs can stack up.

We shared our Welcome Pack last week. (Here is a link to the one we did for our unconference last year). This is an effort we make for any significant sized event we host to be more inclusive - particularly events that aren’t just professionals attending. It seeks to make the space and the structure feel more familiar for people before they arrive.

What I need to take care of

PAIRS - Finish the reimbursement payments, the video clips and the final Programme Committee meeting (5th March).

Our conference - check and double check the equipment I’m taking and finalise a “task list” for team members so we know what we’re doing on the day. Also must up the number on the table booking for the social dinner the night before.

Students and chatbots - We’ve been sub-commissioned to do another piece of work for DfE - this time around student use of chatbots (for educational purposes). The work has an incredible tight delivery schedule required and we’re running two in person workshops (one for students and one for experts) a matter of days apart in March. There’s the usual venue and refreshments booking to handle for both, plus the students will be supported to attend so I’m booking travel from across the country (and wrangling train timetables to do so effectively).

I’m once again in a state of trying not to book a student to end up at our conference and someone attending our conference to end up at the student workshop. Still I seemed to have managed that for India - to my knowledge no-one turned up there expecting to be in Manchester!

It’s March - and that means end of the financial year. So there’s reconciliation tasks related to that, and then liaising with our accountant for the end of year accounts. It also means we need to start our Annual Report but I can’t see myself starting to pull that together until mid-March at the earlier…although perhaps on the train home from Manchester next week (unless there’s post its to be typing up!)

What I’ve been inspired or challenged or moved by

I am an Olympics fiend so I’ve spent the last few weeks watching as much Winter Olympics as the BBC could broadcast. It really is inspirational although it seems to me that most of the sports are dicing with death the whole time, so I’m not sure I’m inspired to give any of them a go specifically!

What I’ve been reading

My reading for pleasure continues at pace. I’m currently enjoying (although the spoiler of knowing how it ends is a bit of a downer) ‘107 days’ by Kamala Harris. I’ve previously read Obama’s post-presidency book, Michelle Obama’s “Becoming”, Hilary Clinton’s book about losing to Trump, and Nick Clegg’s book about the UK coalition government. Kamala’s (if I may call her that) is by far the most engaging and accessible read (in my humble opinion).

I also recently read this article by a CEO I know about managing the absence of a CEO. I found it interesting not so much in the practical elements of the absence (and returning to work) but how organisations need to structure so that there isn’t reliance on any one individual or role (in case it suddenly, or even planned, isn’t there).

Digital Skills

Sat, 14 Feb 2026 00:00:00 +0000

I’m supporting a new open letter on Prioritising AI Literacy for All People.

Last year, just after government announced plans to fund a national AI skills programme, we coordinated an Open Letter to the Secretary of State for Science, Innovation and Technology on the need for an inclusive and critical approach to AI skills and literacy. On behalf of an informal coalition of critical AI literacy experts, we asked to meet with DSIT, and, eventually, had a reply from DSIT offering to arrange a meeting. But, reshuffles intervened, and despite us chasing up, no meeting was arranged.

Which is why it was even more dismaying to how far the launch of the AI Skills Hub in late January missed the mark of what is needed.

Critical data and AI literacy is a foundation for public participation. But delivering this at scale requires the government to listen to and engage with researchers and practitioners who have been developing AI literacy at the front lines.

A number of the co-authors and signatories of last years’ letter have come together to produce an update - open now for signatures - as well as to put the case for critical AI literacy in the media.

Tech Policy Press: The Real Cost of the UK’s ‘Free AI Training for All’ is Democracy
Computer Weekly: The UK government’s AI skills programme betrays UK workers and our digital sovereignty

Emily Weeknotes

Thu, 05 Feb 2026 00:00:00 +0000

Dearest gentle reader…It continues to be a period of heavy workload and trying to balance equally urgent demands across three big events (PAIRS in person, PAIRS online and our own conference). And bonus, we’ve picked up another urgent commission…Jeni’s protecting me from most of that though (as she did from doing the Board papers for next week’s meeting) - it’s appreciated but a tricky feeling.

What I’ve been doing

Before I forget here is a link to a video that is well worth a couple of minutes of your time - it amplifies the voices of students that took part in our pilot of a distributed dialogue around generative AI in education. It serves to remind us all how important it is to involve - and hear - young people in decisions that affect them.

Aside from event logistics (progress is being made but wow there’s still so much being a bit scrabbled around) and my freelance gigs I mentioned in my last weeknotes that I was attending the first Strategy Design Festival, hosted by Dirk Slater and Allen Gunn, which I did.

Pitched as “an opportunity to discover how participatory, adaptive approaches can make strategy a living, breathing practice that your organisation weaves into your ongoing work, and uses to guide real decisions rather than sitting on a shelf” - I was intrigued into how about thinking about this for different organisations I work with and in particular Connected by Data who are committed to participatory practice in what we do. The day was highly interactive (not always a comfortable space for me) and one thing that struck me most was the huge generosity of everyone there. People wanted to listen and show curiosity about each others’ work and strategy experience. Particular shout out to Jess Day and Petra Backovska who are working with fab organisations empowering communities and shared openly with me about their learning. The whole day was well paced and I’ve left with lots to think about - some to enact immediately and some to come back to when it is an appropriate time…and that feels like the best balance of “actions” from a jam packed day.

As a footnote - I want to remember to come back to exploring more about these organisations that were represented there and intrigued me but I didn’t get to chat to them.

As another footnote - but an important one - the event was held at the wonderful Coin Street Neighbourhood Centre. It was my first time here and I highly recommend it as a venue with great spaces and staff, and I plan to book it myself in the future.

Talking of venues, yesterday I “popped” to Manchester for the day to scope out our venue for our March conference. The Mechanics Institute is the “birthplace of the TUC”…and they’re very proud of that association. The building has not been updated so much of the original decor (including a range of stained glass windows marking the different union chapters) remains. I think it’ll be a good space for us.

What I need to take care of

This is less a ‘todo’ and more about how I approach a particular situation. I feel very uncomfortable in small group discussions with people I don’t know (particularly if the subject isn’t something I feel strong in). I was reminded of how hard I found this recently at an event when we were very quickly put into groups where the task meant we all had to talk/contribute. I’m very comfortable learning by being talked at (in an engaging way) and maybe some discussions later. I know that everyone has different learning styles and there’s a desire to make events “interactive” but gosh if that doesn’t undermine the experience for me (and I suspect I’m not alone).

What I’ve been inspired or challenged or moved by

Intrigued more than inspired or motivated … I am a Premier League football fan (for my sins) and have always been interested in tracking how well a professional pundit does at predicting the weekend’s results and how well he does compared to a ‘celeb’ guest. This year BBC Sport have added in an AI predictor - they simply ask Microsoft Copilot Chat to “predict this weekend’s Premier League scores”.

I’m fascinated in seeing how the accuracy of all three pans out over the season (for reference the pundit always beats the celebs over the period of a season even if on occasional weeks they beat him). Over halfway through the season and on a total points basis the AI is winning (although actually is behind ‘readers’ - but I’d argue the volume of these are a major factor). However the leadership table is always done on outright win weeks - which brings the pundit out on top at the moment.

What I’ve been reading

What a revelation reading every day for the National Year of Reading has been. I finished “Tomorrow, and Tomorrow, and Tomorrow” by Gabrielle Zevin in a mere 20 days! It was a really enjoyable read with a long timeline (which I enjoy too). I can see why it was a best seller.

Am now onto ‘Cheer up love: Adventures in depression with the crab of hate’ by Susan Calman.

CWU and AI - Activating members and building towards negotiation

Thu, 22 Jan 2026 00:00:00 +0000

Addressing the complex issues surrounding AI at work can be challenging. It touches on personal experiences, professional roles, and charged political views on Big Tech. It often involves vague or technical terminology. It’s a topic that many employers will see as their ‘management prerogative’ and out of scope for collective bargaining or even worker consultation.

In that context our work with the Communications Workers Union (CWU) – and unions generally – is designed to demystify AI and break down its implications into actionable union responses.

Taking the learnings from Phase 1, in the second phase of our project with the Communications Workers Union, the union led the way to activate its members, while Connected by Data supported the development of a proactive bargaining agenda on AI. The union has launched a web page setting out progress and resources for members, including ‘AI Action Weeks’.

These contributed to strengthening the three part strategy to build CWU power on AI articulated in Phase 1:

Activating and building member confidence

Following an initial five-month action learning project with the telecoms sector executive (including full-time staff and elected members), the union customised the project approach and materials to reach members across a wider range of workplaces.

Several dozen representatives and members attended five sessions. Adam from Connected by Data led the first two, with the remaining sessions led by CWU officers.

The sessions covered:

What does AI look like now? Introducing the current state of AI, core terminology, and examples of its use in daily life (virtual assistants, personalized recommendations, speech recognition, and chatbots). This session also included practical examples of AI’s limitations.
What are the impacts of AI on the world of work? Examining AI in the workplace for tasks such as staff monitoring, hiring, and work allocation. The discussion covered potential risks to members, including job displacement, work intensification, and discrimination, as well as potential benefits such as reduced hours and improved efficiency.
What AI systems, products and software are currently used in CWU workplaces and how are they used? A practical workshop dedicated to identifying specific AI systems, products, and software used across various workplaces. Key discussion points included data collection practices, members’ current knowledge of these systems, and protections under data protection legislation like GDPR.
What control do we have with the implementation of AI in our workplaces? Addressing the legal and collective bargaining landscape of AI implementation. Discussions centered on current legislation, existing collective agreements on AI, strategies for using collective bargaining to ensure union and member involvement in the introduction of AI, and membership education on the topic.
What does the future hold with AI in our workplaces? Focusing on future potential uses, risks, and benefits of AI for members and the union. Specific attention was given to the implications of AI in conduct, grievance, attendance, and productivity issues; strategies for dealing with job displacement; and methods for keeping up-to-date with emerging AI technologies.

Workshopping a model collective agreement on AI

In parallel with activating members, we have worked with the union on clarifying the union’s negotiating position. Unions are in the business of bargaining and making agreements, so you have to know what you want.

We convened leading academics, lawyers, and trade unionists to scrutinize a draft collective bargaining agreement designed to be the basis for CWU negotiations with employers.

The model agreement is founded on three high-level principles that anchor the union’s negotiating position:

Worker voice at all levels of tech decision-making, design, and deployment
A fair share of AI-related benefits including pay, skills, working hours, and training
Robust safeguards against bias, discrimination, and other harms

By drawing on excellent resources, such as UC Berkeley’s inventory of collective agreements on technology, we were able to refine these high-level principles and draft them into the form of a contract between workers and employers.

The model agreement is designed to establish a robust, proactive framework for the introduction and use of any ‘high-risk’ AI system that impacts employees, rather than waiting for the employer to take the initiative.

At its core, the agreement establishes a Joint Union-Management AI Committee (JAIC). This committee features equal representation of worker and employer representatives and binding power, meaning no high-risk AI system can be procured or deployed without the union’s explicit agreement.

This proposed committee sparked significant discussion around potential challenges: Could it undermine the union’s independence? What is the process for a deadlock? How will union representatives be resourced and supported to participate on equal footing with management?

The model agreement also draws on the TUC’s AI Bill to mandate a Workplace AI Risk Assessment (WAIRA for high-risk technologies. Additional provisions address collective data rights and skills.

Working from a preliminary draft, we conducted a critical, multi-step assessment of the draft agreement to strengthen its text and inform strategy:

Scrutinising the model agreement: A detailed review to identify strong provisions and areas needing improvement.
Anticipating employer pushback: A role-play based discussion to anticipate which provisions are likely to be most contentious during negotiations.
Identifying essentials vs. desirables: Pinpointing the union’s “red lines” – provisions considered essential – and other key desires to shape an ambitious and achievable negotiation strategy.

Next steps

Connected by Data is continuing to advise CWU as the union looks towards advancing negotiations across multiple employers. Crucially, the approach demonstrates that unions don’t need to be technology experts to assert control over workplace AI, just as many employers aren’t. What matters is clear principles, structured member engagement, and contractual provisions that establish worker voice throughout the AI lifecycle.

Student Voice and the EdTech Product Safety Expectations

Mon, 19 Jan 2026 00:00:00 +0000

Alongside the Generative AI for Education summit in London yesterday, the DfE updated their ‘Product Safety Expectations’ for EdTech developers, and for schools to consider when deciding which tools are safe to use.

Many of the updates reflect discussions from the GenAI in Education: Have Your Say process - offering a powerful mandate from students for these expectations. Check out what students had to say in the video here https://connectedbydata.org/ai-in-education/ or the full report.

The updates are significant, adding new sections on cognitive development, emotional and social development, mental health and manipulation: and I suspect many tools currently being explored or deployed in schools won’t yet meet the new standards.

The new cognitive development expectations capture the idea of a ‘schools mode’ that pupils raised in GenAI in Education: Have Your Say. DfE say “We expect products not to provide final answers, full solutions, or complete worked examples by default”, but instead to help students learn step-by-step.

Under emotional and social development, the expectations oppose the anthropomorphising of AI tools: calling for tools to drop self-descriptions or conversational modes that might imply products have their own agency, and requiring tools to include time-limits on their use.

Mental health expectations call for tools to detect signs of learner distress, to signpost learners to human help if required, and to involve child mental health expertise in product design.

The expectations on manipulation calls on tools not to deceive or mislead users, to portray absolute or unjustified confidence, or to use peer-pressure to motivate engagement. Tools should not steer users towards paid options through biased wording or layouts.

These should not be radical expectations to have of generative AI tools in schools, and yet, when EdTech tools are built on foundation models tuned towards sycophancy and fast answers over pedagogic principles, implementing them can involve swimming against the tide.

It’s great to see such a strong connection between the views students shared in our distributed dialogues, and the updated product safety expectations. The real test though now is whether these expectations can be operationalised.

That might require one more future update: updating the recommendations on design and testing to strengthen the role of both government and students in making decisions about whether AI for Education tools are ready for use.

We heard across the distributed dialogue that students want to see tools proven as safe, accurate and fair before they are deployed. They expect those assurances from the government, but they also want students to be listened to and involved in those decisions.

Deliberative approaches that provide space for exploring not only whether AI tools do what they say, and meet required standards, but also assessing whether they make a net positive contribution in practice to the kind of education that students and teachers value, should be a central part of this.

Join us for a Webinar on April 22nd when we’ll be talking more about learning from Generative AI in Education: Have Your Say, and thinking about ways for students to have a powerful voice in decisions about data and AI from classroom to the capital.

(Footnote: Because gov.uk pages lack a tracked changes, I had to dig out this archived copy of the standards to fully see what’s changed).

Emily Weeknotes

Fri, 16 Jan 2026 00:00:00 +0000

What I’ve been doing

It is a really busy time at the moment. As a team we’ve got lots of deadlines in the next three months and almost all are involving (to varying degrees) me. I’m feeling really pushed for time and definitely have more than can be done in my working hours. I shared this honestly with Jeni this week and (this will come as no surprise if you know her) she was great - amongst other chatter (and the sharing of a very stress busting photo of her doggy with his chums in the park) we talked through what tasks could be put to one side during this crunch period. I’m not used to having to do that, certainly not in this job, and its tough to accept but also simply has to be.

I reflected that this is where the freelance work I have is tricky. In past times I would have perhaps overworked (and taken the time back). With two freelance roles (who are also busy at the moment) that’s not an option.

What I need to take care of

PAIRS - it’s now just over a month until our full day (extended day to cover timezones from India to East Coast USA) online conference and the next day a full day conference in person in New Delhi. I’m supporting the logistics of both and whilst the ‘on the ground’ bits in India are being done by a partner organisation I’m responsible for managing the travel support grants and fielding enquiries. The big challenge here - nay, stress - is visas. Or more accurately the utter lack of any kind of invitation letters being allowed to be produced but also anyone needing a visa having to have a visa. To make that challenging situation even more “fun” (read “stressful”) is that there’s simply no communication from the Indian Government with any kind of even vaguely helpful suggestions or insights. And this is all linked to a global AI Impact Summit that they’re hosting…there won’t be much global about it at this rate…

The online element is obviously easier from that point of view but we need to make sure multiple rooms running on Zoom, with tens of people in each and moving between, and around 50 presenters and different timezones and tech all comes together smoothly. We’re hoping for volunteers on the day to support with that, but finding / cajoling takes time and effort, as does scoping/articulating the ask and then briefing everyone. … Did I mention it is a month away?! (This of course doesn’t also yet factor in that if the visa issue really doesn’t get sorted then we’ll have a load of other speakers switching to present at the online event if they can’t go to India).

Our conference - in early March we’re hosting a conference on ‘power and participation in public tech’. Hosting a conference is obviously a lot of work. The todo list here includes: planning the space including signage; creating a welcome pack for attendees; getting the invitations out; sorting a team meeting the day before and a ‘get together’ the evening before. There’ll be bits around confirming refreshments too (factoring in dietary needs from the booking attendees) and doing promo on socials for general registration too.

DfE - on Monday Tim and Jeni will attend an AI in education summit hosted by the Department for Education. They will present the outputs of our work using a distributed dialogue process to hear students’ voices on what they think about AI in education. The findings have been interesting and we’ll be sharing those as soon as we can - there’s also a great video which is worth a watch created with a fab group of students from a school in Luton. I’ve got to get that final video on YouTube and available for sharing (once it is finally finalised and once it can be available to share).

Education unions - Adam is doing some amazing work helping education unions develop their thinking and collective bargaining around the use of AI. The unions have developed a statement which is under embargo at the moment but will soon be live (along with a bit more information about the work we’re doing with them).

Next week I’m ‘out of the office’ for a day attending the Strategy Design Festival in London and the week after I’m doing Manchester in a day to suss out the venue for our March conference. That’s more ‘on the road’ (rails more accurately) time than I’ve had in months - I don’t mind it, but it is inevitably less productive time (shonky train WiFi I’m looking at you!).

Time to dig in and get it done…

What I’ve been inspired or challenged or moved by

None of these things but I did giggle at this video where virtual reality (VR) felt a little too real reality for some people.

What I’ve been reading

So this is the most energising bit of my reflections from the last week. 2026 is the National Year of Reading in the UK and my wife and I have committed to making sure we read for pleasure every day. Which means I’m back into reading - and loving it. It really is just about having the discipline to read. I’m currently reading (but have already finished two books this year) “Tomorrow, and Tomorrow, and Tomorrow” by Gabrielle Zevin. I’m at that wonderful place you can get to with a fiction book where I occasionally find myself during the day wondering what is going to happen next and looking forward to reading.