Which Data Sources Should You Care About For AI Search?

TierTypical useSourceEvidence statusWhat the evidence saysReference1Web & search discoveryGoogle SearchConfirmed + currentGrounding with Google Search connects Gemini to real-time web content, returning inline citations to source URLs.Google — Gemini API docs, Grounding with Google Search1Bing SearchConfirmed + currentMicrosoft documents Bing results being used to enhance Copilot responses. Not re-verified in this pass.Microsoft Bing3Common CrawlConfirmed historicalGPT-3 used filtered Common Crawl as roughly 60% of its sampling mixture; LLaMA 1 reported 67%.Common Crawl3Historical web corpora (C4 etc.)Confirmed historicalC4 is a cleaned derivative of Common Crawl; LLaMA reported C4 at 15% of its pretraining mixture.TensorFlow Datasets — C44Web grounding servicesStrong evidence / likelyCategory inference covering third-party grounding/retrieval intermediaries. No single canonical source.—1Products & shoppingGoogle Merchant CenterConfirmed + currentMerchant feed data underpins Google’s shopping surfaces. Retained on user instruction; Google Shopping removed as it is a surface, not a source.Google Merchant Center Help2Merchant / retail feeds (OpenAI)Confirmed + currentMerchants share a secure, regularly refreshed CSV/JSON feed of identifiers, descriptions, pricing, inventory, media and fulfilment so ChatGPT can surface products accurately. Refreshes accepted as often as every 15 minutes.OpenAI Developers — Agentic Commerce, product feeds4Microsoft Merchant CenterStrong evidence / likelyEquivalent commercial feed infrastructure; inferred parallel to Google Merchant Center rather than separately evidenced.—4Marketplace feedsStrong evidence / likelyCategory inference. Shopify catalog data is already integrated into ChatGPT, which supports the pattern.OpenAI Help — Shopping with ChatGPT Search1Local & placesGoogle MapsConfirmed + currentGrounding with Google Maps is a documented tool alongside Search grounding, giving models geospatial context.Google Cloud — Grounding API1Google Business ProfileConfirmed + currentBusiness profile data feeds Google’s local surfaces. Carried from the source table; not separately re-verified.—1YelpConfirmed + currentYelp licenses reviews, photos and business information to OpenAI for real-time local recommendations. Beyond grounding it also drives actions: ChatGPT users can book a table or join a waitlist, and Request a Quote lets users contact providers in-chat. Yelp’s 10-Q confirms it is live.Axios; Yelp blog; Yelp 10-Q FY20264OpenStreetMapStrong evidence / likelyWidely used open geospatial corpus; inferred rather than confirmed for any named model.—4FoursquareStrong evidence / likelyLeft in Tier 4 deliberately: the OpenAI deal is Yelp’s, and no equivalent evidence exists for Foursquare.—4TripadvisorStrong evidence / likelyCategory inference for review/travel data. No confirmed deal identified in this pass.—1Knowledge & referenceWikipediaConfirmed + currentExplicitly present in GPT-3’s disclosed mixture and LLaMA (June-Aug 2022 dumps, 20 languages); also widely used as a live reference/RAG corpus.Wikimedia dumps1WikimediaConfirmed + currentSame corpus family as Wikipedia. Licensing is unusually clear: principally CC BY-SA with attribution/share-alike obligations.Wikimedia dumps4WikidataStrong evidence / likelyStructured entity layer; strongly implied by knowledge-graph use but not separately confirmed.—1Community / Q&A / socialRedditConfirmed + currentThe Google deal gave access to the Reddit Data API for ‘real-time, structured, unique content’, and allows Reddit content to be displayed across Google products — i.e. live grounding, not only training.Tom’s Guide (Google/Reddit deal)2RedditConfirmed + currentSame deal, training side: Google may use Reddit posts to train its AI models and improve services such as Search; reported at roughly $60m/yr. NOTE: Reddit is reportedly weighing whether to renew — treat as unstable.Fortune; Neowin/WSJ on renewal doubt4Social platformsStrong evidence / likelyCategory inference covering platform-wide social corpora.—4Forums / communitiesStrong evidence / likelyCategory inference. Overlaps Reddit but generalised to non-Reddit forums.—1News & publisher contentLive publisher pagesConfirmed + currentReached at inference time via search grounding rather than pretraining; retrieval selection and crawlability govern inclusion.Google — Grounding with Google Search2Licensed publisher contentConfirmed + currentOpenAI has multiple explicit licensing partnerships (FT, Axel Springer, AP, News Corp). Terms differ per partner on training vs grounding vs attribution.OpenAI — FT content partnership2Publisher partnershipsConfirmed + currentAxel Springer’s deal includes otherwise paywalled material in answers; AP licensed part of its text archive.OpenAI — Axel Springer partnership3Historical news corporaConfirmed historicalArchive material absorbed in pretraining; distinct from live licensed access.—2Developer / technicalGitHubConfirmed + currentLLaMA used GitHub’s public BigQuery dataset, restricted to Apache/BSD/MIT projects; The Pile separately includes GitHub. Public visibility is not an open licence.GitHub2Stack OverflowConfirmed + currentNamed in licensing-deal mapping alongside Reddit and Shutterstock as a data platform powering multiple buyers.LLM Pulse — AI content licensing deals mapped2Technical docsConfirmed + currentVendor documentation corpora; widely used but not tied to a single disclosed agreement.—4npm / PyPI registriesStrong evidence / likelyWEAKEST ENTRY IN THE TABLE. Relabelled from ‘package registries’ to name examples. No disclosed agreement or documented retrieval use found — consider cutting.—1Travel & commerce actionsGoogle Hotel Center feedsConfirmed + currentWhen Gemini or AI Mode show hotel options with real-time prices, that data comes from the Google Hotels feed. In Aug 2026 Google added hotel booking inside AI Mode completed with Google Pay, so this is grounding plus actions.TechCrunch — AI Mode travel update4Booking / partner feedsStrong evidence / likelyBooking Holdings and IHG are reported as participants in Google’s agentic booking pilot, which supports the direction but stops short of a documented feed spec.InfosTourisme (IHG/Booking pilot)4OTA / commerce sourcesStrong evidence / likelyCategory inference. OTAs run their own rate feeds into these surfaces.—4Reservation / inventory APIsStrong evidence / likelyCategory inference covering booking/inventory endpoints exposed to agents.—