Newly unsealed court filings in The New York Times' copyright lawsuit against OpenAI and Microsoft allege that senior figures at both companies were aware of efforts to access paywalled content and use copyrighted material to train AI models. Internal documents cite a Microsoft executive describing the practice as an "astonishing theft of unprecedented proportions" and internal data suggesting AI-powered search could reduce publisher traffic by up to 93%.
Conversational AI tools are fragmenting the traditional retail search journey, forcing brands to rethink how they structure product data, create content, and measure visibility. Experts say success now hinges on verified product knowledge graphs, schema markup, and cross-functional teams — not keyword density and backlinks.
A University of Illinois Urbana-Champaign preprint studying 30,000 Reddit AI search answers found that already-upvoted, formally written comments were far more likely to be selected than comments rich in personal experience—and that most first-person language was stripped in the final answers.
Google and DeepMind researchers have introduced Dream-RSI, a technique that allows AI agents to test alternative search strategies by replaying recorded past attempts rather than running costly new computations. Benchmarked on Gemini 3.1 Pro and Gemini 3.7 Flash across eight tasks, the method matched or outperformed baselines while sharply reducing the number of live attempts required.
Google is rolling out an experimental "Dive deeper" feature in Google Discover that presents users with an AI-generated topic overview before linking to publisher articles, prompting concern from content creators and SEO professionals about potential declines in referral traffic.
Regina Moura, head of corporate communications at Roche Pharma Brazil, argues that generative AI has fundamentally shifted the rules of digital visibility: earned media and source credibility now drive brand presence, replacing keyword volume as the dominant currency of search.
A new arXiv preprint testing GPT-5.4's citation behavior found that while top-ranked sources are cited far more often in raw data, deliberately swapping source order produced a much smaller and statistically insignificant effect — suggesting observed position gaps largely reflect content quality rather than rank alone. Structured formatting showed a modest but inconclusive citation lift, and model randomness caused citation decisions to flip in roughly one in seven reruns.
A new Search Engine Journal survey of search professionals finds that organic traffic is declining for many teams while conversions hold steady, and that budget plans are tilting sharply toward generative engine optimization (GEO) even though fewer than one in ten marketers feel confident measuring AI visibility.
Content creators and their agencies are adding answer engine optimization (AEO) and AI search visibility to brand pitches, as conversational AI platforms become a new channel for product discovery. Industry figures from Tinuiti, Deloitte Digital, IZEA, Collectively, and IAB UK say measurement standards are still nascent but interest is growing fast.
A new analysis by Olivier de Segonzac maps the layered data pipeline behind ChatGPT's real-time web answers, revealing OpenAI's proprietary Labrador bots, Bing's image index, and SerpApi — the latter currently the target of a Google lawsuit over unauthorized scraping. A federal judge in September 2026 rejected SerpApi's bid to compel Google to disclose licensing agreements.
Google, Cloudflare, and Microsoft are each piloting distinct systems to compensate publishers whose content powers AI-generated responses, but the three approaches differ significantly in how value is defined, what controls site owners hold, and how transparent the resulting data is.
Google is testing a program that pays publishers when their content meaningfully contributes to AI-generated answers, while simultaneously acknowledging that Search Console reporting for AI search results remains inadequate. Meanwhile, Google lowered the follower threshold for Search profiles to 10,000, and Cloudflare introduced a new setting allowing sites to block AI training without cutting off search crawlers.
