If you’re new to “search,” or among the “old guard,” there’s a hazard of several benevolent of hunt origin myopia. A short-sightedness about what sources really matter for your day-to-day work.
This isn’t new in AI search, but it’s another bruise that’s getting punched again, and again (and again). As AI chatbots and distinct tools akin AI Overviews, Copilot, etc., drag from a additional and additional diverse range of data sources, our job gets a lot harder. And additional interesting!
It’s Hard When You Need To Be Focused On Everything
When item is difficult or challenging, that is the location you have an chance to really get ahead. So this is the chance to really commencement reviewing the distinct data sources that could be preferenced or relied upon by AI tools in the future.
I have built a array of hunt sources and categorised them by how helpful I think they’ll be to you correct now.
Using This Data
| Tier | Meaning |
|---|---|
| 1 | Confirmed + current — RAG / grounding / actions |
| 2 | Confirmed + current — training / licensing |
| 3 | Confirmed historic — pretraining |
| 4 | Strong evidence / extremely likely |
Approach the Tier 1 sources alongside the most involvement – as they’ll more-than-likely be worthwhile. Tier 2 and 3 may be small uncomplicated to accomplish or equal be assured they’ll be beneficial and Tier 4 are extremely likely, but lacking confirmation.
AI Data Sources Reference
| Tier | Typical use | Source | Evidence status | What the evidence says | Reference |
|---|---|---|---|---|---|
| 1 | Web & hunt discovery | Google Search | Confirmed + current | Grounding alongside Google Search connects Gemini to real-time web content, returning inline citations to origin URLs. | Google — Gemini API docs, Grounding alongside Google Search |
| 1 | Bing Search | Confirmed + current | Microsoft documents Bing results being used to enhance Copilot responses. Not re-verified in this pass. | Microsoft Bing | |
| 3 | Common Crawl | Confirmed historical | GPT-3 used filtered Common Crawl as approximately 60% of its sampling mixture; LLaMA 1 reported 67%. | Common Crawl | |
| 3 | Historical web corpora (C4 etc.) | Confirmed historical | C4 is a cleaned derivative of Common Crawl; LLaMA reported C4 at 15% of its pretraining mixture. | TensorFlow Datasets — C4 | |
| 4 | Web grounding services | Strong evidence / likely | Category conclusion covering third-party grounding/retrieval intermediaries. No sole canonical source. | — | |
| 1 | Products & shopping | Google Merchant Center | Confirmed + current | Merchant nourish data underpins Google’s buying surfaces. Retained on person instruction; Google Shopping removed as it is a surface, not a source. | Google Merchant Center Help |
| 2 | Merchant / retailing feeds (OpenAI) | Confirmed + current | Merchants portion a secure, regularly refreshed CSV/JSON nourish of identifiers, descriptions, pricing, inventory, media and fulfilment so ChatGPT can exterior products accurately. Refreshes accepted as frequently as all 15 minutes. | OpenAI Developers — Agentic Commerce, merchandise feeds | |
| 4 | Microsoft Merchant Center | Strong evidence / likely | Equivalent business nourish infrastructure; inferred parallel to Google Merchant Center fairly than separately evidenced. | — | |
| 4 | Marketplace feeds | Strong evidence / likely | Category inference. Shopify catalog data is already unified into ChatGPT, which supports the pattern. | OpenAI Help — Shopping alongside ChatGPT Search | |
| 1 | Local & places | Google Maps | Confirmed + current | Grounding alongside Google Maps is a documented tool alongside Search grounding, giving models geospatial context. | Google Cloud — Grounding API |
| 1 | Google Business Profile | Confirmed + current | Business overview data feeds Google’s local surfaces. Carried from the origin table; not separately re-verified. | — | |
| 1 | Yelp | Confirmed + current | Yelp licenses reviews, photos and endeavor data to OpenAI for real-time local recommendations. Beyond grounding it additionally drives actions: ChatGPT users can publish a array or associate a waitlist, and Request a Quote lets users communication providers in-chat. Yelp’s 10-Q confirms it is live. | Axios; Yelp blog; Yelp 10-Q FY2026 | |
| 4 | OpenStreetMap | Strong evidence / likely | Widely used open geospatial corpus; inferred fairly than confirmed for any named model. | — | |
| 4 | Foursquare | Strong evidence / likely | Left in Tier 4 deliberately: the OpenAI agreement is Yelp’s, and no equal evidence exists for Foursquare. | — | |
| 4 | Tripadvisor | Strong evidence / likely | Category conclusion for review/travel data. No confirmed agreement identified in this pass. | — | |
| 1 | Knowledge & reference | Wikipedia | Confirmed + current | Explicitly current in GPT-3’s disclosed blend and LLaMA (June-Aug 2022 dumps, 20 languages); additionally extensively used as a live reference/RAG corpus. | Wikimedia dumps |
| 1 | Wikimedia | Confirmed + current | Same corpus family as Wikipedia. Licensing is unusually clear: principally CC BY-SA alongside attribution/share-alike obligations. | Wikimedia dumps | |
| 4 | Wikidata | Strong evidence / likely | Structured being layer; powerfully implied by knowledge-graph use but not separately confirmed. | — | |
| 1 | Community / Q&A / social | Confirmed + current | The Google agreement gave admission to the Reddit Data API for ‘real-time, structured, distinctive content’, and allows Reddit satisfied to be displayed throughout Google products — ie live grounding, not lone training. | Tom’s Guide (Google/Reddit deal) | |
| 2 | Confirmed + current | Same deal, training side: Google may use Reddit posts to train its AI models and enhance services specified as Search; reported at approximately $60m/yr. NOTE: Reddit is reportedly weighing whether to renew — treat as unstable. | Fortune; Neowin/WSJ on renewal doubt | ||
| 4 | Social platforms | Strong evidence / likely | Category conclusion covering platform-wide social corpora. | — | |
| 4 | Forums / communities | Strong evidence / likely | Category inference. Overlaps Reddit but generalised to non-Reddit forums. | — | |
| 1 | News & publishing company content | Live publishing company pages | Confirmed + current | Reached at conclusion period via hunt grounding fairly than pretraining; retrieval choice and crawlability govern inclusion. | Google — Grounding alongside Google Search |
| 2 | Licensed publishing company content | Confirmed + current | OpenAI has multiple definitive licensing partnerships (FT, Axel Springer, AP, News Corp). Terms differ per partner on training vs grounding vs attribution. | OpenAI — FT satisfied partnership | |
| 2 | Publisher partnerships | Confirmed + current | Axel Springer’s agreement includes alternatively paywalled matter in answers; AP licensed part of its content archive. | OpenAI — Axel Springer partnership | |
| 3 | Historical news corpora | Confirmed historical | Archive matter absorbed in pretraining; distinct from live licensed access. | — | |
| 2 | Developer / technical | GitHub | Confirmed + current | LLaMA used GitHub’s community BigQuery dataset, restricted to Apache/BSD/MIT projects; The Pile separately includes GitHub. Public visibility is not an open licence. | GitHub |
| 2 | Stack Overflow | Confirmed + current | Named in licensing-deal mapping alongside Reddit and Shutterstock as a data phase powering multiple buyers. | LLM Pulse — AI satisfied licensing deals mapped | |
| 2 | Technical docs | Confirmed + current | Vendor records corpora; extensively used but not tied to a sole disclosed agreement. | — | |
| 4 | npm / PyPI registries | Strong evidence / likely | WEAKEST ENTRY IN THE TABLE. Relabelled from ‘package registries’ to name examples. No disclosed accord or documented retrieval use established — regard cutting. | — | |
| 1 | Travel & commerce actions | Google Hotel Center feeds | Confirmed + current | When Gemini or AI Mode display inn options alongside real-time prices, that data comes from the Google Hotels feed. In Aug 2026 Google added inn booking inner AI Mode completed alongside Google Pay, so this is grounding affirmative actions. | TechCrunch — AI Mode journey update |
| 4 | Booking / partner feeds | Strong evidence / likely | Booking Holdings and IHG are reported as participants in Google’s agentic booking pilot, which supports the direction but stops abbreviated of a documented nourish spec. | InfosTourisme (IHG/Booking pilot) | |
| 4 | OTA / commerce sources | Strong evidence / likely | Category inference. OTAs run their own charge feeds into these surfaces. | — | |
| 4 | Reservation / inventory APIs | Strong evidence / likely | Category conclusion covering booking/inventory endpoints exposed to agents. | — |
Source: Chris Green.
There is a chance this array may date seriously – an occupational hazard of “AI Search” at the moment. But another item you could try is appearance onward to anywhere may be fine sources of data AI providers volition be looking for in the future.
If you’re in a niche or a nation (even) anywhere the complete data sources have small traction (i.e., Yelp isn’t huge in the UK), perchance there are another “big players” you may desire to appearance into, equal if there isn’t a confirmed association yet.
The most meaningful investigation you can do as to which of these data sources are most apt shaping AI results today. That’ll necessitate several activity studying generated results from several queries your customers are apt searching for – appearance for the gaps, or the areas you can get ahead.
As always, any feedback or suggestions welcome. Good luck!
More Resources:
- New Data Finds Gap Between Google Rankings And LLM Citations
- Complete Crawler List For AI User-Agents [Dec 2025]
- Timeline Of ChatGPT Updates & Key Events
This article was initially published on Chris Green SEO.
Featured Image: Gorodenkoff/Shutterstock
