Behind the scenes of our first AI Unconference, in partnership with Manchester's Digital Strategy

To us, running our first Unconference felt a bit like taking a leap of faith. You do all the planning, and then you deliberately take a big step back and let the people in the room take the lead. It's an unusual feeling when compared to something like Camp Digital, which we curate very deliberately and spend a lot of time thinking about complementary themes and sessions for the day. But this also felt like a very exciting prospect, creating something truly participant-led.

This AI, Ethics and Public Engagement Unconference came about through conversations with the team at Manchester's Digital Strategy, where we realised that within our individual networks, we were having a lot of the same discussions about AI, the impact it's having, and the challenges and opportunities it will bring. They’ve been doing thoughtful, people‑led work around AI for some time now, including their work with the People’s Panel for AI and a series of events designed to open up conversations about the real-world impacts and applications of AI. Their focus has consistently been on communities, lived experience, and public trust, and not just the technology itself.

When we talk about co-designing digital solutions, it's just as important that we include emerging tech like AI in these conversations and design work. We decided early on in the planning process that this event should really centre that as its guiding ethos. We wanted to create a space where people with different experiences and understanding could participate equally, uncertainty was allowed, disagreement welcomed, and where people could bring their real questions rather than feeling pressure to be experts or adhere to any particular viewpoint.

We also wanted to connect people doing good work on this subject across various sectors and services. Sometimes half the battle is knowing what other people are working on, and where opportunities to collaborate and share learnings and knowledge can be found. An Unconference felt like a natural fit for that approach.

From the start, this was a genuine and hugely enjoyable collaboration. We worked closely with Sherelle Fairweather and Becky McMillan to shape the format, sense‑check ideas, and think carefully about who we wanted in the room. The aim was to bring together people from public services, charities, tech, research and community organisations, all with different perspectives and experiences. Diversity of voices was key, because the value of sessions like this comes from what people learn from each other.

So, what IS an Unconference?

As the name suggests, an Unconference inverts the usual conference format. There’s no fixed agenda set in advance. Instead, the agenda is shaped by the people in the room, on the day.

Participants suggest sessions based on the questions, challenges or ideas they are already grappling with, and those sessions become conversations rather than presentations. Anyone can contribute, whether that’s by sharing experience, asking questions, or simply listening.

Our role as organisers wasn't to control the content, but to create the right conditions for attendees to contribute. A clear structure, shared expectations, and a welcoming environment where people feel comfortable speaking honestly, even when things feel uncertain or unresolved.

At its best, an Unconference creates space for learning together. It values lived experience as much as expertise, welcomes disagreement, and recognises that not every conversation needs a neat conclusion to be useful. Indeed, a lot of the value comes from starting conversations that hopefully continue long after the event itself has wrapped up.

We took inspiration from the GovCamp model, which has long shown how effective it can be create space for a room full of brilliant people to create their own agenda.

I'm also grateful to GovCamp Scotland organiser Hugh Wallace, who generously shared his thinking and advice from organising their return event last summer. Hugh's advice helped us think about how to create the right conditions to keep things open without losing momentum, and how to make the format feel welcoming, especially for people attending an Unconference for the first time.

Keeping people as the focus

We were also lucky to be supported in running this event by the brilliant team at Digital Futures at The University of Manchester, whose involvement also helped ground the day in ongoing research, practice and public conversation around emerging technologies. Their support reinforced the importance of connecting policy, academia, public services and communities, rather than treating them as separate worlds.

The day opened with an excellent keynote from Professor of Urban Planning and GIS and Director of the Digital Futures Platform, Richard Kingston, who helped set the tone by framing AI as a social and civic issue, not just a technical one. His opening reminded us why these conversations matter, particularly in public services, where decisions shaped by technology have very real consequences for people’s lives.

We also heard from our very own Molly Watt, who spoke about the importance of accessibility, and why it still has to sit at the heart of how we think about AI. Molly spoke about how AI can both remove and create barriers, depending on how it’s designed, implemented and governed. Her talk was a reminder that accessibility isn’t an add on, and that if AI doesn’t work for people with access needs, low confidence or complex lives, then it doesn’t work for public services at all.

Together, these opening sessions helped anchor the day in the needs of people and communities, and set the tone for the conversations that followed.

What goes into running an Unconference?

For this kind of event (as with any), most of the work happens before the day itself.

We spent a lot of time thinking about who we wanted in the space, and why. The aim wasn’t to create a room full of people who already agree with each other, but one where different roles, experiences and perspectives could sit alongside each other. That mix is what makes the conversations richer, and often more challenging in a good way.

On the day itself, our role shifted from organisers to facilitators. Once sessions were proposed and the agenda was shaped by attendees, the focus was on holding space, supporting conversations, and making sure people felt safe and able to speak honestly. Our job was made incredibly easy by all the brilliant people who came along, shared their thoughts generously, and created a fantastic open atmosphere.

Themes that ran through the day

Although sessions covered a wide range of topics, some clear themes came up again and again.

Trust was everywhere

Trust in AI, trust in organisations using it, and trust in how data is collected, stored and used. In many conversations, trust was seen as a bigger barrier than the technology itself, particularly in public services and sensitive contexts.

Human involvement is key

Across discussions about healthcare, social care, recruitment, and education, there was strong agreement that a human in the loop should be the default. AI can support, augment and speed things up, but it shouldn’t replace judgement, empathy or accountability.

Efficiency isn’t the same as quality

Several sessions challenged the idea that faster is always better. Participants talked about the risk of losing context, nuance and human connection in the pursuit of efficiency, and questioned who actually benefits from time saved.

Inclusion and representation are critical

From who builds AI systems, to who gets access to them, and whose data they’re trained on, there was a consistent concern that existing inequalities are being reinforced. Digital exclusion, low confidence, language barriers and accessibility featured strongly across discussions.

Literacy and education are part of the solution

Rather than simply promoting AI tools, there was a clear appetite for better AI literacy. Helping people understand what AI is, what it isn’t, where it works well and where it doesn’t was seen as essential to safe, ethical and confident use.

Pace and pressure create risk

Many participants talked about feeling pressure to adopt AI quickly, often before guardrails, policies or shared understanding are in place. This creates anxiety about getting things wrong, a risk of public backlash, and unintended consequences.

Letting the conversation continue

One of the things we liked most about the Unconference format is that it doesn't try to wrap everything up neatly and then put it away. The questions raised on the day aren’t going away, and in many cases they’re only just emerging.

If you were part of the day, thank you for bringing your experience, honesty and curiosity into the room. And if you’re thinking about running an Unconference yourself, we’d wholeheartedly encourage it... We had a blast!

To help these conversations continue, Manchester’s Digital Strategy team have set up the Tech for Good Community Manchester group on LinkedIn. If you’re interested in tech for good, civic tech or social change in the city, please do join and take part.