What this means
Finding candidate information or documents for an AI response.
Historical record
Google Search upgrades AI Mode with Gemini 4 Hydrogen
A change in how Google AI Search handles complex queries and retrieves links is material to observing source exposure and citation patterns over time.
Google announces planned support for shared AI training preferences in robots.txt
The proposal would allow publishers to express broad training-use preferences without automatically treating them as equivalent to search-crawling or answer-grounding controls. Google says websites that specifically allow Google-Extended are eligible for grounded appearances and clicks in Gemini Apps.
Microsoft announces separate AI training and grounding preferences for Bing publishers
The policy separates traditional search discoverability from permission to inform AI-generated answers. Publishers can make distinct decisions about model training and grounding, while restrictive grounding choices can reduce opportunities for AI citations and brand visibility.
Multi-tool agent retrieval study finds complete tool-set discovery is harder than single-tool matching
Agentic discovery depends on discovering a complete compatible capability chain, not merely ranking a single agent, plugin or tool. This suggests evaluating dependency coverage, completeness and latency alongside tool-level relevance.
OpenAI commits to shared robots.txt AI training preference signal
A provider-specific adoption commitment changes the publisher governance roadmap, potentially reducing separate AI-training opt-out rules. Training permission must be distinguished from search crawler access, grounding and citation outcomes.
ATLAS-Bench evaluates evidence-grounded local search recommendations under user constraints
AI discovery measurement for local businesses needs to distinguish being retrieved from being an acceptable, verifiably supported recommendation. The benchmark highlights a trade-off between strict constraint satisfaction and breadth of acceptable recommendations.
xPRT research trains dense retrievers using relationships across passages
Retrieval quality can depend on whether embedding models learn relationships between passages within documents. This is an upstream determinant of what evidence a retrieval-grounded AI system may find.
Research finds AI news citations more concentrated than audience traffic across countries
AI citation share can concentrate visibility among a narrower set of outlets than traditional traffic, and domain-only measurement may overlook shared ownership. AI discoverability audits can track publisher and owner diversity separately.
Glance study tests generator-based reranking with shared candidate context
AI evidence selection can involve reranking after initial retrieval. Generator-based reranking could affect which retrieved passages inform an answer, independently of whether those passages were initially discovered.
Sequential-decision research shows search policy and retriever jointly shape evidence exposure
AI agents may expose or miss relevant evidence depending on query-branch decisions and allocation of attention across ranked lists, not just the rank of a webpage for a single query.
Medical AI Overviews study separates source selection, citation grounding and accuracy
Shows why AI citations, source credibility and answer-grounding must be measured separately when assessing AI discoverability and reliability.
Research benchmarks citation-ranking models across ten LLM engines under controlled retrieval
Provides a reproducible approach to studying citation allocation after candidate sources are fixed. It distinguishes the source-selection stage from live web retrieval and warns that general relevance scores are imperfect proxies for citations.
Perplexity researchers introduce contextual embeddings for multi-chunk evidence retrieval
AI search depends on finding not only answer snippets but supporting source evidence. Contextual chunk retrieval may change which document passages are available for grounding, though live citation impact is not demonstrated.
Taobao Pailitao reports production gains from multimodal generative product retrieval
Multimodal AI shopping discoverability depends on more than exact product-title matching: user instructions can request visual similarity, changes to attributes, or complementary items. The study connects product retrieval choices with downstream engagement and transaction outcomes in one real commerce ecosystem.
SearchHorizonBench separates answer success from the marginal value of another agent search
A relevant page may not enter an AI agent's evidence pool because the agent decides not to search again. Retrieval-trigger and stopping decisions must be measured separately from ranking and final citations.
ACARe research shows task context can improve reranking of agent-generated search queries
Query fan-out alone may not explain which documents are selected for agentic answers. Context-aware reranking adds a distinct source-selection stage that visibility measurement and retrieval experiments should account for.
OpenAI details agentic retrieval, ranking and auction architecture for ChatGPT Ads
Paid conversational discoverability depends on opportunity detection, semantic retrieval, eligibility, auction and frontend display, each distinct from organic source citation and brand mention. Measurement needs separate funnel stages.
LASAGNA research finds isolated RAG component gains do not reliably combine
AI discovery and source-selection systems need end-to-end retrieval and faithfulness measurements rather than relying on individual component benchmarks or extrapolating citation performance from retrieval accuracy alone.
How AI agents search: Peer-reviewed ASQ study maps agent query reformulation and retrieval workload
AI agents can issue numerous evolving retrieval queries for a single user information need. Measuring only the initial query can miss intermediary source exposure and retrieval workload. This evidence supports treating query fan-out, retrieved source exposure, and final citations as separate stages.
Research shows evolved semantic identifiers improve controlled generative retrieval recall
Semantic identifiers and collision handling influence which documents or products are retrieved in generative search/recommendation systems, independently of downstream answer wording or citations.
OpenDS-World benchmarks evidence discovery by agents in live web environments
AI discovery depends on whether agents can find, navigate and interpret source material before deciding what to cite or use in an answer.
Arcade study measures complete tool-set discovery and agent session cost
Agent discoverability requires retrieving the entire set of tools needed to complete a task, not just some individually relevant APIs. Catalog scale, authorization, reranking and token cost jointly shape which actions an agent can discover and perform.
GLOBE evaluates reusable, field-scoped product-listing optimization for generative shopping
AI shopping discoverability involves structured product titles, attributes and descriptions; controlled ranking feedback can help assess whether rewriting improves relevance exposure or introduces downside risk.
Controlled research finds location channels change AI citations and business recommendations
AI visibility and GEO audits that use a single geographic origin can mischaracterize brand and source exposure. Declared location, network country and run-to-run variation need to be recorded as separate measurement variables.
Retrieved but Not Read study separates search-result retrieval from source reading and citation faithfulness
Source retrieval and even citation do not guarantee that an AI answer read or faithfully used the underlying evidence. This adds a measurable distinction between discovery, source opening, evidence use and answer support.
SPRA research introduces source-linked structured paper objects for scientific agents
Agentic evidence discovery can depend on whether a system retrieves isolated passages or source-linked objects that preserve methods, findings and their conditions. The work supplies measurable evidence for structured scientific retrieval.
AURA research finds adaptive reasoning depth improves retrieval relevance and latency
The amount and direction of query reasoning can change which documents are selected for AI answer generation. This adds retrieval-utility and stopping-policy dimensions to source selection, query fan-out and AI discovery measurement.
Cloudflare–OpenAI pilot uses network signals to expand ChatGPT Search index discovery
AI search discoverability can involve separate URL discovery, index admission, retrieval, citation and referral stages. The pilot makes upstream network discovery a relevant measurement dimension, without demonstrating a citation or ranking boost for participating sites.
Apple study finds tool discovery should differ for single-tool and multi-tool agent tasks
AI agents can fail to discover required tools even when their task intent is understood. Different discovery architectures expose different capabilities, so agent visibility measurement should distinguish direct named-tool lookup from multi-step tool-set discovery.
Glasseek research introduces auditable deep-search agent evaluation
Agentic search benchmarks can overstate retrieval capability if an agent encounters hidden answer keys or benchmark data while searching. The work highlights why evidence provenance, search trajectories and benchmark leakage must be audited before drawing conclusions about AI discovery.
Research traces ChatGPT Web retrieval-to-citation behavior across model settings
AI citation measurement cannot reliably infer all cited URLs from Google Search or API proxy results alone. Repeated sampling and separating retrieval from citation outcomes across model settings may improve GEO measurement validity.
Research separates retrieval, content matching and model knowledge in AI citations
AI visibility measurement should distinguish retrieval eligibility and generated subqueries from answer citation and brand mentions; formatting-only interventions cannot be assumed to change selection.
SEEDS research separates retrieved candidates from evidence admitted into agent reasoning
An AI agent may retrieve a page yet never admit its full content into the reasoning context. AI discovery and citation measurement should distinguish retrieval exposure, preview inspection, evidence admission and final answer citations.
Google documents a dedicated fresh UGC ingestion pipeline for Search
It documents a gated upstream Search discovery and freshness pathway for first-hand UGC. This is relevant to source availability but not evidence that AI Overviews, AI Mode or Gemini will cite submitted pages.
Research demonstrates citation-governance risks from low-barrier web publishing
Citation presence alone is not proof of editorial trust; source-selection and publication governance can create measurable evidence-integrity risks.
Perplexity releases multimodal late-interaction retrieval models
The work shows a concrete retrieval design that preserves token-level and visual evidence instead of compressing each document into one vector. For AI discoverability, source retrieval fidelity and final citation selection must be evaluated separately.
IndexAct research separates candidate-set refinement from text inspection in agentic search
Agentic discovery can depend on how an agent narrows candidate sources and when it chooses to inspect evidence, not only on which documents a search index initially matches.
T-Search research separates multi-step evidence retrieval from answer generation
AI discovery has an upstream evidence-retrieval stage distinct from final citation selection. Multi-round searches and alternative retrieval rollouts can change which source chunks are available to a generator; citation monitoring alone does not measure this upstream coverage.
Agentic retrieval study quantifies quality gains alongside steep latency and token costs
Source discovery in multi-step agents can differ from a single semantic similarity ranking, but additional query fan-out and reasoning impose material latency and cost. Evaluation of AI discovery pipelines should track evidence quality and search budget together.
Programmatic Search Agents demonstrate that evidence presentation matters beyond query reformulation
For AI discovery, a page entering retrieval is not equivalent to its supporting passage reaching the model or being used in an answer. The research supports separating retrieval, evidence delivery and final source use in visibility and attribution measurement.
Cloudflare launches AI Gateway Web Search API with source-link requirements
Adds an intermediary retrieval layer for agents and a source attribution requirement for partner web search results. This affects how retrieval infrastructure can expose content, independently of consumer AI answer citations.
RAAC study finds deep-research agents repeat searches after evidence discovery saturates
In agentic discovery, repeated search calls do not necessarily expose new documents. Retrieval stopping policies and marginal evidence coverage can affect which sources an agent sees before composing an answer.
MCite-RL research improves visual citation grounding through iterative evidence refinement
A document or image can be retrieved without its specific supporting region being cited accurately. Multimodal AI discovery requires distinguishing retrieval, region selection, citation localization and answer support.
EvidenT study tests source verification before enterprise RAG answer generation
AI answer citations can drift from retrieved evidence. Source-grounding verification is a distinct stage between retrieval and final citation display, relevant to trustworthy discovery measurement.
EBG study finds complete evidence retrieval and citation coverage are distinct RAG bottlenecks
AI discovery pipelines can retrieve a source without gathering enough passages to support a multi-part answer. Answer-only evaluation and raw citation counts can miss incomplete grounding.
EMNLP study traces deep-research citation errors to agent orchestration
Citation visibility can fail after retrieval: evidence can be discovered and propagated between agents yet lose attribution during orchestration or synthesis. This makes citation survival distinct from initial source retrieval.
DailyReport evaluates web research agents with task-level factuality and user-oriented rubrics
AI discovery measurement cannot stop at retrieved links or citation counts: agents must find, verify and synthesize evidence into responses satisfying the actual user task.
Cloudflare makes AI Search generally available with native multimodal retrieval
The release provides a concrete implementation of how parsing, multimodal indexing, retrieval and reranking govern which evidence reaches downstream answer generation in developer-built systems.
EMNLP study finds deep-research agents can mistake correlated pages for independent corroboration
Citation and source counts can exaggerate corroboration when multiple retrieved pages repeat a common origin. Agentic discovery measurement needs to distinguish independent evidence, copied evidence, and which documents the retriever actually exposed.
MiRAGE introduces claim-level information and citation metrics for multimodal RAG
As AI answers draw from video, image, audio and text, the number of citations or text overlap alone cannot establish whether a particular claim is supported by the cited media. Citation support and information coverage require separate measurement.
ReCite evaluates intent-aware retrieval and verification for faithful academic citations
Authentic and topically relevant retrieved sources can fail to support a particular claim. The work provides a testable approach to distinguishing candidate retrieval from faithful citation selection.
VHop-Router study moves multi-hop visual evidence search into the retriever
Visual and multimodal AI discovery can be limited by a retriever's ability to traverse evidence relationships, not just by the reasoning model. Evidence-chain completeness and search cost are distinct metrics from initial top-k matching.
Peer-reviewed e-commerce search study finds hybrid retrieval more consistently improves relevance than dense-only search
AI shopping discovery must satisfy explicit attributes as well as semantic intent. Retrieval pipelines should be evaluated for query-type relevance and not assume dense semantic matching is universally superior to keyword matching.
INSPIRE benchmark measures exposure, selection and ranking losses in research agents
A source that an agent never retrieves cannot be cited, and retrieved evidence can still be omitted or poorly ranked. Citation visibility measurement should distinguish these failure stages.
Perplexity describes Photon in-house retrieval and ranking engine
The architecture shows how index representation, candidate retrieval, ranking and latency affect the source-discovery pipeline before an answer is generated.
Tool-Lab study links costly product-attribute retrieval to shopping-agent choice errors
AI shopping discoverability includes whether agents inspect relevant product attributes, not only whether a product is retrieved or recommended. Product data accessibility and instruction specificity may affect which evidence is used during selection.
Structured semantic supervision research distinguishes intent-satisfying products from merely related listings
Shopping discovery depends on whether a product actually meets the query's primary requirements, not just textual similarity. Structured attributes and explanations offer measurable signals for product selection.
Cross-engine GEO audit finds little exact-URL citation overlap in a fixed-prompt sample
Cross-platform AI visibility dashboards need to distinguish page/query fit, observed source exposure, conditional citation selection and final citation presence, rather than assuming a platform-neutral page score predicts mentions.
Google rolls out UCP integration hub for native checkout on AI Mode and Gemini
AI shopping discoverability is extending beyond whether a product is retrieved or recommended into whether merchant commerce capabilities are machine-readable, eligible and executable on an AI surface.
Research characterizes cross-platform conversational web search behavior
The results provide empirical evidence that AI discovery is a multi-stage process: deciding to search, formulating machine-written queries, retrieving candidate sources, selecting evidence, grounding responses, and exposing citations can vary independently across platforms.
Google Search Live moves to Gemini 3.8 Live
Search Live is an AI discovery surface that exposes web links during conversational voice interactions. A model-layer change creates a new measurement boundary for source selection, citation exposure, follow-up behavior, and multimodal discovery experiments.
Google Research demonstrates property-aligned query fan-out with Retrieve-for-Train
It provides primary research evidence that AI retrieval can depend on complementary semantic coverage across a fan-out, rather than only matching a single canonical query, and gives Visibility OS a measurable framework for thinking about fan-out quality.
Perplexity introduces CobbleDB for lower-latency, lower-cost AI search storage
The work provides first-party evidence about how an AI answer engine is optimizing the infrastructure beneath retrieval, helping separate content-level GEO factors from systems-level retrieval constraints.
Perplexity rebuilds AI-search storage around prepared passages, versioned document state and query-time serving
This gives direct first-party evidence that AI-search discoverability operates through multiple infrastructure stages between crawling and answer generation, including document preparation, versioning, export policy, retrieval/ranking and passage selection.
Two structured GEO experiments test earned-source effects on AI visibility
The experiments provide reproducible hypotheses for Visibility OS around source selection, citation concentration, source decay, entity co-occurrence, and the difference between owned-content optimization and earned-source visibility.
Perplexity introduces Q2D-Web for large-scale retrieval evaluation in agentic RAG
The visible user prompt is not necessarily the query a source competes for. AI visibility therefore depends on several distinct stages: query reformulation, first-stage retrieval, ranking, evidence selection, and eventual citation.
Gemini 3.8 Flash rollout briefly disrupts citations in Google AI Mode
For GEO measurement, citation loss should not automatically be interpreted as loss of retrieval eligibility or ranking. Model and answer-generation changes can alter whether retrieved evidence becomes a visible citation.
Research demonstrates defenses against malicious Generative Engine Optimization
AI visibility interventions operate inside source-selection systems that may change defensively. A tactic that increases citation or source use in one evaluation may not remain effective if retrieval or reranking systems learn to identify optimization patterns as manipulation.
Counter-GEO-Bench measures defenses against information-distorting GEO
The benchmark makes adversarial source selection measurable and distinguishes conventional safety filtering from defenses designed specifically for information-distorting GEO.
CHASE study finds repeated GEO adaptation can weaken ranking–quality alignment in simulation
It identifies a potential long-run failure mode of optimizing publisher content exclusively against an observable ranking proxy: improving ranking-feature fit need not preserve source quality, which matters for evidence selection and citation governance.
ITER uses prior agent search history to improve iterative evidence retrieval
Multi-step agentic discovery is path-dependent: the same current subquery can be evaluated in light of earlier questions and visited evidence. Source visibility measurement should consider complete search trajectories.
ARD v0.91 defines JSON-LD agent discovery and the ard.json manifest path
Publishers relying solely on the earlier manifest path risk being invisible to conforming ARD consumers; the proposal also separates semantic relevance ranking from identity and trust verification.
GEO-Flag research measures GEO-like web content and citation-source verifiability
AI visibility measurement now has a proposed way to audit whether retrieved source content resembles optimization-driven rewrites and whether its outbound citation URLs are verifiable, rather than equating citation frequency with source quality.
WANDR benchmark evaluates wide-and-deep research agents
Agentic discoverability is not only a single-query ranking problem. A source or entity may need to be found during broad discovery, revisited during enrichment, and retained in an evidence chain. WANDR provides a checkable framework for measuring those stages together.
Research models GEO citation competition and tests verifiable-content rewards
The experiments make a concrete distinction between improving source factual value and optimizing superficial citation cues. They illuminate how competitive GEO could degrade answer quality in repeated interactions.
Breadcrumbing study shows adaptive search-result injection can steer multi-step agent evidence chains
Agentic retrieval and source selection can be influenced over multiple search steps, so corroboration among search results does not automatically establish independence when an intermediary can alter the observation stream.
SIEVE demonstrates structure-aware search, inspection and section-level fetching for research agents
Agentic discovery may depend on which parts of a source can be addressed and fetched, not just whether the page appears in a search index. Structured document access affects retrieval precision and context efficiency in the tested system.
EMBL AI Librarian demonstrates multi-query evidence retrieval in scientific research agents
This provides direct research evidence for separating multi-query retrieval, source selection and passage-level evidence grounding when analyzing AI discovery pipelines; it does not measure public-web publisher visibility.
Large-scale study separates citation-pool presence from entity exposure in generative search
AI visibility measurement should not collapse retrieval, citation and entity exposure into one metric. Interface should also be treated as an experimental variable when auditing generative-search visibility.
SearchOS research shows explicit evidence coverage and shared search state improve benchmarked agent discovery
Long-horizon AI research agents may fail to expose sources or assemble grounded answers when they lose track of missing facts or repeat searches. The work gives a testable framework for measuring discovery completeness, fine-grained evidence provenance and search-state management beyond simple retrieval or citation counts.
NASA Earth Observation benchmark demonstrates hybrid retrieval and agentic reranking
Specialized resources such as scientific datasets can become more discoverable through structured knowledge-graph metadata, hybrid retrieval and task-aware reranking rather than simple keyword matching alone.
LoRE study reports improved e-commerce search relevance from multimodal and rule-aware reasoning
AI shopping and product discovery require correct matching of user constraints to item attributes, images and business rules. This research provides production-affiliated evidence that relevance adjudication is a distinct stage influencing which products can surface.
Google announces Agentic Resource Discovery for federated agent-capability discovery
Agentic discovery gains a distinct search and verification layer for callable services, separate from conventional web-page indexing and generative-answer citations.
EComAgentBench evaluates shopping agents on hidden user intent and multi-step product verification
AI shopping discoverability cannot be reduced to single-query product retrieval: final selection may depend on whether an agent acquires, remembers and correctly applies user preferences and review-based constraints.
XBCP research finds language mismatch weakens agentic retrieval and citation fidelity
AI discovery of non-English sources can fail at multiple stages: retrieving relevant evidence and faithfully integrating that evidence into a cited answer. Visibility and citation metrics need language-aware, stage-specific evaluation.
FORGE study finds one polluted retrieved page can induce fake-product recommendations
A single retrieved source can contaminate generated shopping recommendations, so source admission, source provenance and product-identity validation are distinct from citation visibility.
SafeGEO finds GEO source rewrites can promote flawed recommendations despite partial defenses
AI shopping and agentic product discovery can be distorted when seller-controlled source text misrepresents product fit. Recommendation audits must evaluate hard constraints and evidence support, not just whether a product is mentioned or cited.
RISE research tests bounded, navigable evidence workspaces for agentic search
For AI discovery research, the experiment distinguishes document candidate inclusion, access to evidence inside a retrieved document, and final answer generation. It demonstrates that evidence-delivery interfaces can influence agent effectiveness and cost in controlled tasks.
Mechanistic study separates citation-marker generation from evidence use in RAG
Citation presence or correctness is not equivalent to demonstrating that retrieved evidence causally informed the answer. This distinction matters for measuring AI visibility, source selection and grounded answer quality.
Deep literature search study separates citation coverage from topical relevance
AI discovery measurement should separate whether sources were found from how they were ranked and judged relevant; existing human citations are not a complete or unambiguous ground truth for evaluating retrieval.
LLM-Wiki research shows agent-guided traversal of linked evidence can improve multi-hop retrieval
The experiments distinguish first-pass document matching from agent-directed traversal of intermediate entities and evidence. This matters to AI discoverability measurement, source provenance and multi-hop information access, but does not establish a commercial AI search ranking or citation benefit.
Controlled SIGIR study isolates competitive first-citation selection across six LLMs
AI visibility analysis should distinguish retrieval and candidate ordering from subsequent citation selection; content relevance, price detail and recency can matter once a source is in the candidate context.
ACM health-search audit finds limited overlap between Google AI Overview citations and organic sources
AI citation visibility and organic search ranking are distinct measurements. Site owners and researchers need to measure AI Overview activation, citation share, source overlap and platform differences rather than infer them from classic SERP ranks.
Research shows repeated UGC retrieval can contaminate deep-research citations
Multi-query agentic retrieval can repeatedly select the same editable sources. Citation occurrence alone is not evidence of factual reliability; provenance, moderation and source integrity affect trustworthy AI discovery.
SGR-Bench identifies website retrieval-state failures in AI search agents
AI discoverability can fail even when an agent finds the correct website: answer-bearing records may remain inaccessible or be interpreted under the wrong site-specific data context. This suggests measuring source discovery, state configuration, evidence extraction and final answer quality separately.
Longitudinal audit measures AI Overview activation, citation overlap and claim grounding
Conventional first-page rankings, AI Overview triggering, citation inclusion, and claim fidelity are different visibility and quality measurements; one should not be substituted for another.
SIRA demonstrates corpus-aware one-shot retrieval as an alternative to repeated agent search
Agentic source discovery does not necessarily require more query fan-out. Discriminative vocabulary and corpus-aware retrieval controls may change the efficiency and source-selection behavior of future retrieval-augmented systems.
Direct Corpus Interaction research challenges fixed top-k retrieval for search agents
AI evidence discovery may depend on an agent's ability to inspect raw documents, combine constraints and revisit candidate sources, not only on similarity-ranking scores. This adds a distinct interface design to retrieval and source-selection research.
Peer-reviewed audit finds AI Overview citation selection differs by document provenance after rank controls
AI discoverability measurement should separate observed retrieval position from citation selection and record document provenance. Organic rank alone cannot be treated as an AI citation metric.
SIGIR research finds substantial source divergence between Google Search, AI Overviews and Gemini
AI discovery measurement must distinguish organic Google rankings, AI Overview citations and Gemini results, repeat query measurements and document crawler policy separately.
Peer-reviewed study measures attribution gaps between retrieved and cited AI-search sources
It distinguishes retrieved source exposure from visible citation credit, revealing limitations in using citations alone to infer publisher contribution.
AgentSearchBench evaluates agent discovery against actual task execution
As AI agents become discoverable services, matching an agent's description to a user request may not reliably identify the agent that can complete the task. The benchmark offers a method to distinguish description-level exposure from execution-grounded selection.
FeatGEO research evaluates feature-level optimization for citation visibility and content quality
AI citation visibility and content quality are not interchangeable. This work suggests document-level properties can be evaluated jointly with citation outcomes, while requiring controlled testing rather than assuming individual rewrite tactics work universally.
MAGEO research tests reusable GEO strategies with fidelity-aware citation evaluation
Provides a controlled methodology for separating content-edit effects from retrieval-list changes and for measuring whether increased exposure is faithfully attributed, rather than optimizing citation counts alone.
CoSearch jointly trains agent reasoning and evidence ranking for multi-step search
Intermediate agent-generated subqueries and the document admission/reranking stage can shape which evidence reaches a reasoning agent. This suggests retrieval and reranking should be evaluated separately from final citations.
Cloudflare enforces canonical redirects for verified AI training crawlers
Separates the governance of model-training crawlers from live search crawlers and browsing agents, giving publishers more precise control over which version of content is fetched for training.
Google brings side-by-side webpage exploration and tab context to AI Mode in Chrome
Browser-integrated AI discovery combines algorithmic web retrieval with user-selected page and file context, creating a different route by which publisher content enters an AI conversation.
MERRIN benchmark exposes multimodal evidence-selection and over-search failures in web agents
Multimodal AI discoverability requires that agents retrieve and correctly interpret the right source modality, not merely find a webpage. More query fan-out can introduce distracting evidence and impair final grounding.
Peer-reviewed study finds synthetic web content can amplify retrieval exposure
Source diversity, synthetic-content exposure, and retrieval integrity can degrade without immediate drops in answer accuracy, complicating the measurement of trustworthy AI discovery.
Google explains Googlebot's 2 MB fetch cutoff and indexing consequences
Critical text, canonical metadata and structured data located beyond Googlebot's fetch cutoff may be invisible to Google Search indexing. This is a concrete upstream discoverability constraint for oversized pages.
LRAT trains search retrievers on agent browsing and reasoning trajectories
Documents that help an agent resolve an intermediate information gap may differ from those most relevant to the initial user query; retriever training can therefore change which sources are surfaced to AI agents.
Ctrip study finds shopping AI chat and conventional search often interleave
Shopping AI discovery can coexist with conventional keyword search during the same purchase journey; AI visibility measurement should track both channels and intent progression.
Gemini hotel citation audit finds query intent shifts OTA versus non-OTA source shares in Tokyo
AI citation exposure in intermediary-heavy categories can depend on whether a query seeks a bookable attribute or a traveler experience. Visibility measurement should segment by intent, language and source class, without equating a grounding citation to traffic or booking.
Research finds AI-search source selection and answer synthesis diverge across systems
AI visibility measurement must distinguish source discovery, citation appearance and source contribution to synthesized answers. Observed source pools and synthesis patterns differed by tested system; a single citation count or traditional Google ranking cannot represent every AI discovery surface.
Shopping Companion benchmark links long-term preference memory to agentic product selection
Agentic product discovery can fail after product retrieval if the agent forgets user preferences, skips attribute checks, or mishandles budget and bundle constraints. This provides a benchmark for separating retrieval availability from recommendation validity.
Controlled RAG study finds enhanced linked entity pages outperform plain HTML and JSON-LD alone
The study directly tests whether machine-readable entities and navigable relationships can improve evidence retrieval and answer completion. It distinguishes adding schema markup from providing richer linked entity pages.
AgentGEO research diagnoses and repairs citation failures across simulated generative-engine stages
Shows why AI citation measurement should distinguish source access, extraction and final citation decisions; content changes cannot solve every upstream or domain-level selection problem.
MC-Search benchmark measures stepwise multimodal retrieval and agent planning
Visual and agentic AI discovery can require multiple dependent searches across text and images. Measurement of evidence discovery needs hop-level and modality-aware diagnostics, not just final-answer accuracy.
ProductResearch trains shopping research agents using supervised web and catalog trajectories
Shopping-agent discovery can involve both open-web source selection and structured catalog retrieval, with evidence coverage and product attributes shaping research reports rather than just first-pass item ranking.
KDD research finds text-only GEO edits can harm upstream retrieval in full-pipeline tests
GEO tests that hold retrieved sources fixed can miss upstream discoverability losses. Publishers and researchers should separately measure retrieval inclusion, reranking position and answer citation rather than assuming content already enters model context.
SAGE benchmark finds agent query wording changes retrieval effectiveness
Which sources enter an AI agent's evidence pool depends on its intermediate query formulations and retrieval configuration; semantic similarity alone does not predict exposure across tasks.
Pinterest reports production GEO framework for visual-content acquisition
For AI discovery, the study provides production-scale evidence that content representation, semantic aggregation, retrieval architecture, and internal authority distribution can be engineered together. It also shows that GEO measurement can be tied to acquisition outcomes rather than citation counts alone.
Large-scale agentic search logs reveal cross-step evidence reuse and intent-dependent query trajectories
Agentic discovery involves sequences of queries informed by prior retrieval, so measurement limited to the initial prompt or a single cited result can miss how sources enter later searches.
Research compares generative-search citations with Google source overlap and freshness
The observed differences show why Google search position cannot be treated as a direct proxy for AI citation visibility, especially when source categories and freshness differ by intent and system.
Agentic-R trains retrievers for multi-step search using local relevance and final-answer utility
In multi-step AI discovery, passages selected for semantic similarity to a subquery may not help complete the final answer. Retriever training objectives can therefore change which evidence reaches an agent, independently of traditional search ranking or publisher content changes.
E-GEO introduces an e-commerce benchmark for generative engine optimization
Provides an attributable framework for evaluating product-listing visibility in conversational shopping instead of relying on untested rewriting rules.
Study compares language predictability and semantic similarity in generative-search citations
It adds original evidence that citation selection and answer synthesis need separate measurement from organic search rank and that source-text properties can correlate with generative citations.
Large-scale GEO study finds AI-search source selection varies by engine, language and query phrasing
Visibility measurement should preserve engine, language and query-variant dimensions rather than collapsing them into one visibility score. Source type can also be measured as an observed selection outcome, while avoiding unsupported claims that it is a universal ranking factor.
DeepTRACE audits citation reliability across generative search and deep-research systems
AI visibility measurement can overstate source quality when it counts citations without checking whether the cited source actually supports the generated statement. Reliable measurement therefore needs citation correctness and support, not only citation frequency.
RAID study tests GEO rewriting without access to future search queries
It distinguishes optimization for an unknown range of future information needs from tuning a page for a known prompt. The benchmark encourages evaluation across multiple unseen query variants and separate measurement of source use versus real-world retrieval.
Bing connects sitemap lastmod and IndexNow to AI-powered search discoverability
Crawl coverage and freshness can make updated content retrievable in Bing-powered AI experiences, but do not guarantee AI citation.
Study finds model-specific concentration in AI search news citations
Provider-specific citation concentration and diversity can differ substantially. AI visibility measurement should track source distribution and publisher diversity separately for each model family instead of relying on pooled citation totals.
Google expands AI Mode and Deep Search using query fan-out
Topical completeness, subtopic relevance and retrieval across multiple intents become central to AI search visibility.
Anthropic launches web search on the Claude API
AI discoverability extends beyond consumer search interfaces into embedded enterprise and developer workflows.
Google introduces AI Mode and describes query fan-out
Visibility is no longer tied only to the user’s literal query; sources can be discovered through hidden subqueries and related retrieval paths.
ALCE introduces an end-to-end benchmark for LLM answers with citations
Citation frequency alone does not establish answer verifiability; evaluation must assess whether references actually support generated statements.
ChatGPT web browsing begins rolling out in beta
Website discoverability became directly relevant to ChatGPT answers and later evolved into a dedicated search product.
OpenAI WebGPT demonstrates browser-based answering with citations
The retrieve-read-cite pattern closely anticipates later AI answer engines and modern citation optimization concerns.
Retrieval-Augmented Generation (RAG) is formalized in research
RAG is one of the core technical patterns behind modern AI search, source retrieval and citation behavior.