The Squeeze Tightens: As AI Search Drains Traffic, Newsrooms Cut Deep and Sue AI Firms

Two forces are converging on the news business in 2026, and together they are reshaping it. Artificial intelligence is draining the search traffic that funds journalism — and publishers, with fewer options left, are fighting back in court.

The clearest symbol of the squeeze came from one of America’s most storied newsrooms: The Washington Post cut roughly a third of its staff this year, including more than 300 of its approximately 800-person newsroom, its deepest reductions in years. Leadership pointed to a near-halving of organic search traffic over three years as AI reshaped how people find information.

The traffic collapse

For two decades, the deal underpinning online news was simple: publishers wrote stories, Google sent readers, and some of those readers became subscribers. AI is breaking that chain.

Google’s AI Overviews — the summaries that answer a query at the top of the results page — increasingly mean users never click through to the source. Data cited by Digiday linked AI Overviews to a roughly 25% drop in publisher referral traffic, and a majority of major publishers report search-traffic losses somewhere between 1% and 25%. When the answer appears on the results page, the “search-to-subscription” funnel that paid for reporting simply stops working.

The layoff wave

The cuts are not isolated, and they are not only American. The BBC told staff it plans to shed about 2,000 jobs — roughly one in ten. Politico trimmed around 3% of staff to start the year. The Atlanta Journal-Constitution cut about 15%. The Associated Press reduced editorial staff by some 60 positions through buyouts and layoffs. In Europe, the publisher Bauer cited Google’s AI Overviews and falling click-throughs as it restructured digital operations in Germany and the UK.

The pattern is transatlantic and structural — a response not to a single bad year but to a business model under siege.

Lawyer up

The other half of the story is litigation. On May 28, 2026, CNN filed its first lawsuit against an AI company, suing the answer-engine startup Perplexity for copyright infringement after licensing talks broke down. CNN joined a growing crowd: Perplexity now faces roughly six publisher copyright suits, including from The New York Times, the Chicago Tribune and News Corp.

The legal theory is consistent across the cases — that AI firms are building products on the back of copyrighted journalism without permission or payment, and that “zero-click” summaries substitute for the original work rather than pointing readers to it.

Sign up — or get scraped

Running parallel to the lawsuits is a licensing economy. Some publishers have struck deals to be paid for the use of their content by AI developers, and proposals for statutory licensing — which would force AI firms to pay for news much as radio stations pay for music — are gaining traction. The industry is, in effect, pursuing two strategies at once: sue the companies that take content without paying, and sell to the ones willing to.

The strategic reckoning

The deeper shift is in how newsrooms think about success. For years the metric was pageviews, and pageviews depended on search. As that source erodes, publishers are pivoting toward audiences they own directly — newsletters, apps, subscriptions — and toward direct deals with the AI platforms now mediating between readers and the web.

It is a painful transition, measured in lost jobs and courtroom filings. But the alternative — continuing to depend on a search funnel that AI is quietly dismantling — is no longer viable.

Sources

1. Poynter — Washington Post layoffs (2026)
2. Deadline — CNN sues Perplexity for copyright infringement (May 2026)
3. Digiday — Google AI Overviews linked to ~25% drop in publisher referral traffic (2026)
4. Press Gazette — Journalism job cuts in 2026 (running tracker)