Google.org is backing a new Jigsaw Partner Program that will let more local governments use Sensemaking AI to collect and analyze large volumes of public feedback. The program is designed to help officials turn open-ended conversations with residents into policy insights, with participating localities able to run the work at no operational cost through partner organizations.
Google announced the expansion on September 22, 2026. Jigsaw, a Google incubator, introduced Sensemaking AI about a year ago as a set of open-source tools for public leaders. The tools combine AI-assisted interviewing with automated analysis and reporting. Instead of asking every resident the same short list of survey questions, the system can ask follow-up questions based on a participant’s earlier answers, then organize the resulting responses into themes and areas of agreement.
That approach is meant to address a familiar weakness in civic consultation: traditional surveys are good at producing comparable numbers, but they often leave little room for people to explain why they hold a view. Town halls can provide richer discussion, but they reach only a small slice of the population and can favor people who have the time, confidence, or local connections to attend.
Jigsaw’s pitch is that AI can make the qualitative part of public consultation operate at a much larger scale. Its Sensemaking AI site describes four main components: adaptive interviewing, automated reporting, predictive agreement, and panoramic insights. The first two are relatively easy to understand. The system asks personalized follow-up questions, then groups and summarizes the responses. Predictive agreement is intended to identify statements that may receive broad support based on what participants have already said. Panoramic insights add local public data, research, and examples from peer cities to help officials assess possible solutions.
The program is not starting from a blank page. Jigsaw says a month-long conversation in Bowling Green, Kentucky, gathered input from 8,000 participants for a 25-year growth plan. Residents submitted more than 4,000 proposals and over one million opinions, according to the project’s own account. The new partner program will extend similar work to cities and states including Chattanooga, Houston, Bentonville, Riverside, Kitchener, San Rafael, Schenectady, and Gdańsk. Jigsaw also says it is working with the National Governors Association and the Napolitan Institute to bring Sensemaking AI to five additional states.
The expansion will be delivered through a network rather than as a simple software download. Jigsaw says Change.org, Make.org, and RMG Research will act as platform partners. They can help recruit participants, design a conversation around a specific policy question, run the engagement, and turn the results into reports for officials and the public. Google.org, Bloomberg Philanthropies, and the David & Lucile Packard Foundation are supporting the collaborative fund behind the program.
Jigsaw’s software remains open source, which gives developers and civic-technology groups a way to inspect, fork, and deploy the core tools themselves. But the partner model also acknowledges that running a large civic conversation involves more than code. A government still needs outreach, participant recruitment, translation, data handling, moderation, and a process for deciding what to do with the results.
That operational layer is where the important questions begin. Jigsaw says 94% of participants in its pilots felt their opinions were accurately represented, compared with 47% for traditional surveys. It also reports high satisfaction figures from individual projects. These are useful signals about the product’s goals, but they are vendor-reported results, not an independent evaluation of every deployment. The outcomes could vary with the questions asked, the participant pool, the language used, and the way officials interpret the final report.
There is also a difference between hearing more voices and giving those voices decision-making power. An AI system can summarize a conversation, surface recurring themes, and find areas of agreement. It cannot by itself guarantee that a city will act on the findings, that minority views will not be buried in a summary, or that participants understand how their data is processed. Jigsaw’s FAQ says data privacy is managed according to each platform partner’s policies, which means residents will need to look beyond the Jigsaw brand to understand the exact data practices in a particular project.
For PeDeDe readers, the interesting part is not simply that Google is putting AI into public consultation. It is the product shape: a model that treats conversation as structured data, uses an AI interviewer to gather more detail, and then presents officials with a map of public priorities. That could make participation easier to scale, but it also makes the quality of the questions, the transparency of the analysis, and the accountability of the officials more important.
Google.org and Jigsaw are inviting local leaders to apply for the program. The first wave is planned across 30 cities, states, and countries over the next year. Whether Sensemaking AI becomes a useful civic layer or just another dashboard will depend on what happens after the summaries are produced: who can inspect them, how disagreements are preserved, and whether public input changes an actual policy decision.
Sources: Google.org, “Using AI to help local governments connect with constituents” — https://blog.google/company-news/outreach-and-initiatives/google-org/sensemaking-ai/ Jigsaw, “Sensemaking AI” — https://jigsaw-code.github.io/sensemaking-tools/ Jigsaw, “Helping civic leaders hear every voice” — https://medium.com/jigsaw/helping-civic-leaders-hear-every-voice-5704646900d6 Jigsaw, “One year on: The momentum behind Jigsaw’s Sensemaking AI” — https://medium.com/jigsaw/one-year-on-the-momentum-behind-jigsaws-sensemaking-ai-b340e89ed516