Collective Intelligence Network Get involved

Get involved

For civil society

When a new technology reshapes daily work, everyday people should not be left on the outside looking in. Member-owned groups can build and test practical safeguards today.

Long before modern regulations existed, neighbours joined together to solve problems that no single family could handle alone. Farmers built grain elevators to bypass rail monopolies. Factory workers formed mutual aid funds to care for injured friends. Families pooled small savings into credit unions when commercial banks turned them away.

Today, artificial intelligence is reshaping how people find housing, earn a living, and receive healthcare. Automated tools now screen tenant applications and set rent prices across whole neighbourhoods. Algorithms manage delivery shifts and score job applicants before a human manager ever sees a resume. If each person faces these systems alone, they have no leverage to ask how decisions were made.

Civil society exists to balance that scale. Co-operatives, credit unions, worker associations, tenant organisers, and community clinics hold a unique power. They already have the trust of their members and the legal standing to pilot collective safeguards. They do not need to wait for federal rules to protect their communities.

Comparison of isolated individuals versus civil society collective coordination Diagram showing how isolated individuals face opaque algorithmic pricing and screening without leverage, while civil society coordination pools member data trusts, private multi-party audits, and democratic deliberation to safeguard community interests. ISOLATED INDIVIDUAL No leverage TENANT SCREENING Black box applicant filtering Single applicant cannot audit scoring DYNAMIC PRICING Algorithmic rent & utility fees Hidden markups split across tenants WORKPLACE PLATFORMS Automated shift quotas Individual accounts face take-it-or-leave-it terms CIVIL SOCIETY ACTION Shared power COMMUNITY AUDIT CLINIC Joint algorithmic denial audits Turns individual rejections into legal evidence PRIVATE SET MATCHING Blinded bulk buying & audits Compares supplier bills without opening books MEMBER DATA TRUST Collective bargaining on AI Sets binding bounds on workplace monitoring
How community institutions turn isolated vulnerability into collective oversight.

Four practical safeguards your organisation can test today

1. Member-governed data trusts

Most commercial software gathers member data and uses it to train proprietary models without consent. A data trust changes the terms of custody. Your organisation holds data under collective governance rules decided by members. When an outside provider requests access, the trust enforces strict boundaries. Personal records stay scrubbed, and findings flow back to the community.

Recent events show this collective leverage works. In November 2026, a major platform tried to claim AI training rights over every website linked on its network. Coordinated pushback from creators, publishers, and civic groups forced the company to pause the rollout.5 When communities organize together, platforms have to listen.

2. Private matching to catch hidden overcharges

Small clinics and independent businesses often face price markups that larger chains easily negotiate away. Until now, small buyers could not easily pool orders without exposing their private accounting books. Modern privacy-preserving matching lets several groups discover shared suppliers and compare bulk needs safely. No organisation reveals its raw finances or internal client lists. The math reveals only where groups can join together to demand fair rates.

3. Community algorithmic audit clinics

When an automated system denies housing or sets shifts, the reasons often stay locked inside a proprietary box. Community clinics can run structured audits on behalf of residents. By gathering anonymised denial letters or shift schedules, organizers can spot systematic patterns. These clinics turn scattered complaints into rigorous evidence for regulators and courts.

4. People-centred deliberation assemblies

Automated tools should never decide high-stakes community policies without public consultation. Deliberation panels bring frontline workers, residents, and domain specialists together to review proposed tools. Participants set clear red lines before software touches public funds or member services. This keeps democratic accountability firmly in human hands.

How to begin

You do not need an engineering staff to take the first step. Pick one process where your members feel pressure from automated decisions. Map what records you hold, what questions you want answered, and which neighbouring groups share the same challenge.

We provide open research specs, test frameworks, and briefing materials designed for non-profit and community teams. If your group wants to explore a pilot, read our open scenarios in Shared Horizon or write to us at contact@collectiveintelligencenetwork.org.

  1. Ostrom, E. (1990). Governing the Commons: The Evolution of Institutions for Collective Action. Cambridge University Press. https://doi.org/10.1017/CBO9780511807763 ↩

  2. Freedman, M. J., Nissim, K., & Pinkas, B. (2004). Efficient Private Matching and Set Intersection. In Advances in Cryptology — EUROCRYPT 2004, LNCS 3027, pp. 1–19. https://doi.org/10.1007/978-3-540-24676-3_1 ↩

  3. Council of Europe. (2024). Framework Convention on Artificial Intelligence and Human Rights, Democracy and the Rule of Law (CETS No. 225). Vilnius, 5 September 2024. https://rm.coe.int/1680afae3c ↩

  4. Berdoz, F., Rugli, L., & Wattenhofer, R. (2026). Can AI Agents Agree? arXiv preprint arXiv:2603.01213. https://arxiv.org/abs/2603.01213 ↩

  5. Belanger, A. (2026). Meta Abandons Sketchy Commercial Terms Update. Ars Technica, 6 November 2026. https://arstechnica.com/tech-policy/2026/11/does-meta-think-we-cant-read-meta-abandons-sketchy-commercial-terms-update/ ↩