226 will be the year one AI search company builds a truly fair value exchange with the publishing industry into its model.
At the moment, AI search is primarily extractive. Bots collect sites and answer search queries in a box that users never leave. In rare but positive situations, publishers may be compensated for initial scraping and given assurance that content will be published accurately, ethically, and with fair attribution. But the old model of the internet, where nodes like Google and Facebook route readers back to publishers and creators, doesn't yet exist in AI. Few publishers actually acquire readers from AI search.
There are many ways to solve this problem. External links from the search box can become more visible. AI companies can identify user intent, and even particularly high-value queries, and direct specific users to the site where they originally provided the information. Larger companies may adopt a model like the one ProRata has built, where publishers are compensated based on their contributions to answers to specific queries. Advertising inventory on AI search sites may be provided based on the value of the information collected. And these marketplace features could be rolled out with stronger guardrails for publisher attribution, factual accuracy, and journalistic ethics.
So far, AI companies have not adopted these ideals because there is no incentive to do so. They are in a race for the limitless wealth and power of AGI and don't want to slow down for a second. But a healthy publishing industry is good for the AI industry because it means better, more accurate information flowing into models. And next year, one of the leading AI companies plans to figure it out.
