Perplexity, the AI search engine that summarises web content in real time, faced a predictable backlash three weeks ago when news outlets discovered their work being stripped and condensed without traffic flowing back. The Forbes and Wired complaint stung because it hit the core vulnerability of these platforms: they collapse the difference between reading and researching. Why click through if the model answers your question in thirty seconds?
The startup's response was pragmatic rather than defensive. They tightened their summarisation algorithm to leave content partially opaque, making clicks necessary. They made attribution explicit. But the real move was financial: a new Publishers Program that reverses the usual extraction logic. Instead of outlets paying to appear in results, Perplexity now cuts them in on ad revenue generated when their content surfaces in responses. Time, Der Spiegel, Fortune, Entrepreneur, and the WordPress ecosystem signed on immediately. This isn't a licensing deal or an exclusivity cage—it's closer to a revenue-share network where multiple sources can participate simultaneously.
It's a smarter play than the exclusives we've seen elsewhere (Reddit with Google, The Atlantic with OpenAI), which force publishers into winner-take-all bets on platforms that may not survive. Perplexity's model hedges that risk. The timing helps too: while the US Department of Justice pursues Google's search monopoly, here's a plausible alternative ecosystem forming in real time. Whether it generates meaningful income for publishers, or merely launders the same extraction problem with a revenue line attached, remains untested. But the architecture—paying sources rather than choking them—is the kind of thing the search market should have figured out years ago.