Civic AI
Part 1: How can AI and collective intelligence enable new forms of community response to the climate crisis?
A version of this piece was published here, in collaboration with Nesta’s Centre for Collective Intelligence Design.
To address the climate crisis, we need to compliment radical policies with an increased capacity for communities to organise and adapt to a new reality. Doing this requires better tools and methods for mobilising large groups of people to take action, reducing associated costs, and advancing the value of collaboration. In this piece, we describe three future use cases that bring together groups of people and AI to address environmental issues at a community level.
Over the last three months Dark Matter Labs and Lucidminds, with support from Nesta, have been developing three use cases for near future ideas where AI can help equip communities with the tools to collectively respond to the climate crisis and achieve the 2050 target of a carbon-neutral economy. This builds and expands on some of our current projects at Dark Matter Labs, including Trees as Infrastructure and Smart Commons, as well as previous work done by Nesta such as The Future of Minds and Machines: How artificial intelligence can enhance collective intelligence.
The AI opportunity
Organising large scale community responses can be a messy and complicated task, but AI can help cut through this complexity to accelerate the coordination of action. We have been developing an Augmented Collective Intelligence (ACI) framework which provides a design tool for applying human and machine intelligence collectively, to respond to community based challenges.
The framework represents augmented collective intelligence as emerging from the interaction of people, technology and objects in the physical world and digital environment. Humans and civic assets (for example a smart solar panel) make up the physical context. Digital traces of humans and machines, combined with simulations and software, make up the digital environment. The framework allows the user to model the interactions between these various components, as well as data and information flows. It explores how communities might harness the collective knowledge and behaviour to manage complex systems, the technical aspects of which will be explored in more detail in a subsequent post.
Nesta’s previous work highlighted at least eight ways that AI can add value to CI. Applying the adapted ACI framework (see Figure 1) we have identified four key areas of opportunity that relate most directly to challenges faced by projects working on climate crisis responses:
- Fostering collective understanding, using machine learning and citizen science techniques to allow for mass data collection and analysis — for example the Solar PV Nowcasting project uses satellite imagery and short-term weather forecasting to accurately predict electricity production from solar panels to reduce the need for backup supply from fossil fuels.
- Recognising that to transition from a human-centric to an ecocentric perspective, we must embrace a more-than-human notion of agency. Including non-human life forms, such as trees; or inanimate things, such as rivers; as well as the fundamental rights of future generations, could be one approach to achieving this. In turn, unlocking a more balanced economy which incorporates gifting, caring and financing for all things. As Jones and Cloke note in Non-Human Agencies: Trees in Place and Time, “the agency of materiality and in this case the materiality of nature, needs to be factored into social explanations of the functioning of the world and its detailed space-time patterns. To ignore the agency of the non-human or merely to talk of social nature, is to help ensure that ‘the world is rendered as an exclusively human achievement in which ‘nature’ is swallowed up in the hubris of social construction’ (Whatmore, 2003: 165)”.
- Recommendation and feedback systems to encourage and sustain collective climate actions through analysing complex issues and recommending optimised actions — for example CityMatrix helps simulate the impact of urban planning decisions to enable collaborative real-time decision making.
- Modelling and visualising multidimensional impacts to provide evidence for outcomes-linked climate response investment — for example Regen Network is developing a platform for automating the contracting of ecosystem regeneration.
Where AI and CI could make a difference
Informed by climate response strategies in Project Drawdown and the Exponential Roadmap, we identified eight possible use cases, related to these three areas. The ideas ranged from community composting networks monitored by distributed sensors and AI assisted modelling to help plan and initiate community-scale retrofit of buildings to developing generative design tools for collaborative urban planning of zero-carbon communities.
Developing potential use cases
There are exciting opportunities for innovation in all of these areas. However, when weighing up their potential impact with plausibility in a community setting over the next three to five years, we believe the most immediate opportunities are in three areas:
- Accounting for the benefits of urban trees
- Collective climate action
- Community energy
Use case 1 — Accounting for the benefits of urban trees
Cities are struggling to match the scale of tree planting needed to meet their net-zero targets, partly because we struggle to measure the vast benefits trees provide us, so we see them as a cost rather than as an investment. What if we use machine learning to partly automate the process of mapping trees and measuring their benefits to justify upfront investment? What if AI agents prompted citizens to verify certain datasets or suggested that nearby trees need watering and their fruit can be picked?
Use case 2 — Collective climate action
Responding to the climate crisis will require radical changes to how we live, but sustained change requires actions which are locally appropriate and guided by an understanding of the benefits of taking action (and consequences of not taking action). How might AI help build shared understanding from diverse perspectives; use feedback to reinforce collective actions; and run simulations to identify the gap between the potential impact of local actions and what is required to meet national climate targets?
Use case 3 — Community energy
Community energy plays a vital role in decarbonising the national grid and boosting local economies. Yet since government support ended in 2019, the sector has struggled, relying heavily on volunteers to set up, administer and operate the projects. Could AI be used to help automate site identification; enable smart ownership contracts; provide remote fault diagnostics and simulate potential lifetime revenue and social impact, in order to reduce the risk of upfront investment and subsequent operating costs?
Next steps
Behind the ideas described in this piece sits a more in depth analysis and description of the technical framework which we will publish over the next few weeks. We recognise that both artificial and collective intelligence can be complex terms to engage with, which poses a challenge to making the research accessible and useful to the organisations and communities already working on climate crisis responses. To overcome this hurdle, we will be taking the next month to develop the three use cases while engaging with civil society organisations (CSOs) to identify how AI and CI can best help them tackle some of their challenges. We will update the ACI framework as we go, to ensure it provides a useful toolkit.
If you’re interested in engaging with the ideas or the technical Augmented Collective Intelligence framework, we’d love to hear your thoughts. Get in touch @Dark MatterLabs.
Update: Read the follow-up to this piece in Civic AI Part 2: An exploration of Augmented Collective Intelligence
Thanks to all the valuable insights we have heard so far, from a number of organisations, including: Brixton Energy, Calthorpe Community Garden, Cambridge Canopy Project, Participatory City, Possible and Scottish Natural Heritage.
Project team
This piece has been co-authored by:
Fang-Jui Chang & Oliver Burgess — Dark Matter Labs
Bulent Ozel, Oguzhan Yayla and Sander van der Hoog — Lucidminds
Visuals: Fang-Jui Chang, Hyojeong Lee & Oliver Burgess — Dark Matter Labs
The project is supported by Nesta’s Centre for Collective Intelligence Design
References
Whatmore, S., 2003, Handbook of Cultural Geography. In Culturenatures: Introduction: More than human geographies, edited by K. Anderson, M. Domash, S. Pile, and N. Thrift, pp. 165–167. Sage, London.
