How Village Media is Building a Moat Against AI and Platforms

Hi, I’m media innovation journalist Ulrike Langer and you’re reading News Machines. If this newsletter was forwarded to you becoming a subscriber means you will never miss an issue. If you want to support my work please spread the word, become a paid subscriber or click on the ad.

A year ago, News Machines covered Village Media in a piece about Spaces, the Reddit-like community layer the Canadian chain launched in Sault Ste. Marie, Ontario. This year at the International Journalism Festival IJF26 in Perugia I sat down with Richard Gingras, Google's former Vice President of News and now chairman of Village Media's board, for a longer conversation about where the Canadian company is heading.

The short version: Village Media is moving into the United States, preparing to launch a first international partner outside North America and sharpening its argument for how a local media company should position itself against AI and platform dependency. The expansion is the news. The argument is more interesting.

Last year Gingras called Village Media a community impact platform with news as a component. This year he calls it a community operating system. The reframe is not only rhetorical. Civic engagement - the workshops, the stakeholder meetings, the partnerships with local actors - is now explicitly positioned as the company's marketing budget. It does good things for the community, Gingras said, and it builds visibility with local advertisers, who supply the revenue.

The clearest expression is the Community Impact Protocol, a new initiative Village Media is testing in Sault Ste. Marie. According to CEO Jeff Elgie on LinkedIn, the local website SooToday polled readers on the city's most pressing issues. The majority picked opioid addiction and mental health. Village Media's civic engagement team then convened stakeholders - health-care providers, first responders, city leaders, non-profits, educators - for a workshop at Village Media's headquarters. Two months later, Gingras told me, the team ran a larger community conversation with around a hundred residents. SooToday is now producing a weekly series on the opioid epidemic: how the community got here, what is working elsewhere, what could be implemented locally.

Gingras credited the underlying idea to Dean Baquet (the former executive editor of the New York Times). The point, in Gingras's framing, is not to solve the community's problems but to help residents understand and address them. This is real solutions journalism - often talked about at journalism conferences but seldom put into practice yet. 

Growth in Canada continues. Village Media expanded to Dundas, its 27th Ontario community, last year after city leaders approached the company and asked them to come in. 

Gingras said Village Media will launch in 15 U.S. communities over the next year. All will be owned-and-operated by Village Media this first phase - a deliberate choice, he said, because the model needs U.S. case studies. His worry is that American publishers currently tell themselves, "That's Canada. Things are different in Canada. People are nicer in Canada." To drive change across the playing field, he said, you need to drive envy. 

Village Media is also about to launch its first international partner outside North America, in the English-speaking Global South. Gingras declined to name the country or the partner, saying they had not yet announced any of this.

Alongside the expansion, Gingras said a large technology company - not Google - has funded Village Media's work on the community operating system but he did not name the company or the amount. 

GenAI is now cross-cutting at Village Media, but Gingras is precise about where. Three uses stand out.

Engineering. Gingras said AI has completely changed how the company builds software. Village Media is now developing more applications across the business because it can.

Market analysis for the U.S. rollout. Village Media hired a consultant to identify regions, then fed the company's market-selection criteria and its historical data to Gemini, which produced further analysis. The team then passed Gemini's output to Claude for critique and sometimes ran the exercise in reverse. Gingras described it as having a team of McKinsey consultants who are not arrogant and do not cost a lot of money. (He acknowledged that the consultants can also be sycophantic.)

An audience-facing product. Most interesting, Village Media is building a concierge product called Open Door that connects residents to local resources for issues such as drug addiction, food insecurity or starting a small business. In Gingras's description, the goal is to simplify people's ability to find local solutions. He was clear that the tool depends on local knowledge that comes from boots on the ground. "You want to make sure you understand the organizations that you're referring people to," he said.

That is the opposite of the approach by Axios Local which is scaling into new local markets with heavy AI use. Gingras is not dismissive, but he is skeptical about where algorithmic scale runs out. "You can scale that way," he said, "but you're only going to go so far." Village Media's model, in his telling, depends on human relationships - a newsroom that understands local stories, a civic engagement team that builds relationships, an advertising sales effort that works with local merchants. AI is an assist, not a substitute.

Gingras was relaxed about LLMs crawling Village Media's daily news feed. He was not relaxed about the rest of the stack. Event directories, classifieds, the Open Door concierge, and other commercial assets sit behind a moat. In his words: you could crawl what is in the daily feed, but an event directory is not going in there.

His logic on licensing follows from this. Gingras is openly skeptical that licensing news content to LLMs will be meaningful revenue for local publishers. The LLMs will pick and choose, he said, and they will go with the big publishers over the small ones. News content, he argued, does not have the commercial value people assume. Big companies are not licensing Gemini or ChatGPT for the news content; they are licensing them for many other things. Where he does see licensing potential is in the curated local layer - an Open Door concierge that other LLMs could book into through a licensing arrangement. He was careful to position this as a supplement that could grow, not a pillar.

Readers familiar with Google's decade-long argument about where the value sits in online advertising will recognize the through-line. Gingras pointed out that even in Google Search, there are no ads against serious news content - advertisers want travel, health and the other niches with commercial intent. That is not a new insight, he conceded. But it is one that, in his experience, the news business still has not absorbed.

The most striking part of our conversation was Gingras's view of what a reporter should be in an AI-saturated news environment. He said he had been talking with David Walmsley, editor of the Globe and Mail, about reorienting newsroom thinking. The old division was reporters who could write and reporters who could report. The new question, in Gingras's framing, is who is good at gathering information - all kinds of information - that AI cannot easily grab. 

That is where the moat is, at the newsroom level and at the company level: the things that can only be learned by being there.

Gingras referred to his own research in Chicago on local information habits. Two findings: 

Gingras’ findings illustrate why traditional, broad-reaching metropolitan newspapers struggle in the digital era - they often fail to provide the hyper-local, neighborhood-level coverage that residents desire, while simultaneously being less essential for national news.

These observations underpin Gingras's argument that successful local news models must prioritize community-building and connection to the immediate interests of their audience rather than attempting to serve as broad, general-interest outlets.

Village Media is now developing a price-comparison tool for its communities. Stumptown Savings, an independent site in Portland, Oregon, is doing something similar. Bryan Vance, the publisher, recently told me the people who benefit most from the service are not necessarily able to pay for it - a reminder that building something genuinely useful is not the same as building something that pays.