Minerval

Browse

Claims

Search the graph by meaning. Each result carries its current verdict; open one to see its decomposition, provenance, and the reasoning behind the assessment.

ShowingImportancePrizesTopicArtificial intelligence32 claims

The study and development of intelligent systems capable of performing tasks that typically require human intelligence, including reasoning, learning, perception, and decision-making.

The 2026 AI mathematical results mark a phase transition in AI models' research mathematics capability.
SupportedEvidence favors the claim, but the chain is incomplete or the sources are secondary.constitutionevaluativeA judgment of worth or quality against some standard: good, fair, effective.constitutionAI mathematical discoveryLLM capability trendsimportance · majorImportance 0.65, from 0 to 1 · major: real consequence within a domain, actively argued. Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution
An AI system has autonomously produced research mathematics of a quality publishable in a leading journal
UnassessedNo current assessment. Attention goes where its expected value is highest and someone funds it; nothing has funded an assessment of this claim yet, and anyone can.constitutionevaluativeA judgment of worth or quality against some standard: good, fair, effective.constitutionAI mathematical discoveryMathematicsimportance · notableImportance 0.50, from 0 to 1 · notable: a contested point in a live debate (also the default before judging). Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution
Frontier AI models produce novel mathematical results through reasoning comparable to a human mathematician's rather than brute-force search
ContestedCredible evidence or argument exists on multiple sides.constitutionempirical · derivedA factual claim that rests on inference from other evidence rather than direct observation.constitutionAI mathematical discoveryMachine insight versus memoryLarge language modelsimportance · notableImportance 0.55, from 0 to 1 · notable: a contested point in a live debate (also the default before judging). Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution
AI labs do not disclose how many failed attempts precede their announced mathematical results
UnassessedNo current assessment. Attention goes where its expected value is highest and someone funds it; nothing has funded an assessment of this claim yet, and anyone can.constitutionempirical · derivedA factual claim that rests on inference from other evidence rather than direct observation.constitutionAI Lab TransparencyAI mathematical discoveryMetascienceimportance · minorImportance 0.40, from 0 to 1 · minor: narrow or largely settled, cheap to get right. Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution
An AI model that generates a verified mathematical result under human direction merits primary discovery credit
ContestedCredible evidence or argument exists on multiple sides.constitutionnormativeA claim about what should be done or how things ought to be, settled by argument rather than evidence alone.constitutionAI discovery creditMathematicsimportance · notableImportance 0.50, from 0 to 1 · notable: a contested point in a live debate (also the default before judging). Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution
AI models' mathematical advantage over human mathematicians comes from encyclopedic knowledge and patience rather than deeper insight
UnassessedNo current assessment. Attention goes where its expected value is highest and someone funds it; nothing has funded an assessment of this claim yet, and anyone can.constitutionempirical · derivedA factual claim that rests on inference from other evidence rather than direct observation.constitutionMachine insight versus memoryAI mathematical discoveryimportance · notableImportance 0.45, from 0 to 1 · notable: a contested point in a live debate (also the default before judging). Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution
Published chain-of-thought records of AI mathematical discoveries show goal-directed strategy rather than exhaustive enumeration
UnassessedNo current assessment. Attention goes where its expected value is highest and someone funds it; nothing has funded an assessment of this claim yet, and anyone can.constitutionempirical · derivedA factual claim that rests on inference from other evidence rather than direct observation.constitutionMachine insight versus memoryAI mathematical discoveryimportance · minorImportance 0.40, from 0 to 1 · minor: narrow or largely settled, cheap to get right. Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution
Large AI-generated developments of formalized mathematics are inevitable.
UnassessedNo current assessment. Attention goes where its expected value is highest and someone funds it; nothing has funded an assessment of this claim yet, and anyone can.constitutionevaluativeA judgment of worth or quality against some standard: good, fair, effective.constitutionAI mathematical discoveryFormal proof verificationimportance · minorImportance 0.40, from 0 to 1 · minor: narrow or largely settled, cheap to get right. Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution
AI language models can find counterexamples to long-standing open mathematical conjectures.
UnassessedNo current assessment. Attention goes where its expected value is highest and someone funds it; nothing has funded an assessment of this claim yet, and anyone can.constitutionevaluativeA judgment of worth or quality against some standard: good, fair, effective.constitutionAI mathematical discoveryLarge language modelsimportance · notableImportance 0.60, from 0 to 1 · notable: a contested point in a live debate (also the default before judging). Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution
Discovery credit can meaningfully be attributed to a non-human AI system
UnassessedNo current assessment. Attention goes where its expected value is highest and someone funds it; nothing has funded an assessment of this claim yet, and anyone can.constitutionempirical · derivedA factual claim that rests on inference from other evidence rather than direct observation.constitutionAI discovery creditimportance · minorImportance 0.40, from 0 to 1 · minor: narrow or largely settled, cheap to get right. Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution
AI coding assistants substantially speed up developers on many programming tasks.
SupportedEvidence favors the claim, but the chain is incomplete or the sources are secondary.constitutionempirical · derivedA factual claim that rests on inference from other evidence rather than direct observation.constitutionAI Developer ProductivityGenerative AI productivity effectsimportance · notableImportance 0.60, from 0 to 1 · notable: a contested point in a live debate (also the default before judging). Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution
The rapid growth in frontier AI training costs will continue through the 2030s
ContestedCredible evidence or argument exists on multiple sides.constitutionempirical · derivedA factual claim that rests on inference from other evidence rather than direct observation.constitutionAI Training Compute Cost Trendsimportance · notableImportance 0.55, from 0 to 1 · notable: a contested point in a live debate (also the default before judging). Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution
Coding benchmark scores and anecdotal reports overestimate real-world AI coding capability.
SupportedEvidence favors the claim, but the chain is incomplete or the sources are secondary.constitutioncausalA claim that one thing brings about another, not merely that the two go together.constitutionAI Coding Benchmarksimportance · notableImportance 0.60, from 0 to 1 · notable: a contested point in a live debate (also the default before judging). Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution
The compute cost of the most expensive AI training run will reach about $230 billion (one percent of 2021 US GDP) around 2040.
ContestedCredible evidence or argument exists on multiple sides.constitutionempirical · derivedA factual claim that rests on inference from other evidence rather than direct observation.constitutionAI Training Compute Cost TrendsTraining compute trendsimportance · notableImportance 0.45, from 0 to 1 · notable: a contested point in a live debate (also the default before judging). Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution
Scaling up training compute will remain the primary driver of frontier AI capability gains
UnassessedNo current assessment. Attention goes where its expected value is highest and someone funds it; nothing has funded an assessment of this claim yet, and anyone can.constitutionempirical · derivedA factual claim that rests on inference from other evidence rather than direct observation.constitutionLLM capability trendsimportance · notableImportance 0.60, from 0 to 1 · notable: a contested point in a live debate (also the default before judging). Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution
AI development and investment are directed mainly at automating existing tasks rather than creating new tasks and products
UnassessedNo current assessment. Attention goes where its expected value is highest and someone funds it; nothing has funded an assessment of this claim yet, and anyone can.constitutionempirical · derivedA factual claim that rests on inference from other evidence rather than direct observation.constitutionAI development directionimportance · notableImportance 0.45, from 0 to 1 · notable: a contested point in a live debate (also the default before judging). Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution
Regulators should require the most capable foundation models to meet quality and safety standards.
UnassessedNo current assessment. Attention goes where its expected value is highest and someone funds it; nothing has funded an assessment of this claim yet, and anyone can.constitutionnormativeA claim about what should be done or how things ought to be, settled by argument rather than evidence alone.constitutionFrontier AI model regulationimportance · notableImportance 0.60, from 0 to 1 · notable: a contested point in a live debate (also the default before judging). Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution
AI capabilities will advance to the point of fully automating most occupations
ContestedCredible evidence or argument exists on multiple sides.constitutionempirical · derivedA factual claim that rests on inference from other evidence rather than direct observation.constitutionHuman-level artificial intelligenceAutomation and employmentimportance · majorImportance 0.75, from 0 to 1 · major: real consequence within a domain, actively argued. Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution
Hard-to-learn tasks lack the objective outcome measures AI models need to learn them effectively.
UnassessedNo current assessment. Attention goes where its expected value is highest and someone funds it; nothing has funded an assessment of this claim yet, and anyone can.constitutionempirical · derivedA factual claim that rests on inference from other evidence rather than direct observation.constitutionTask outcome measurabilityimportance · notableImportance 0.50, from 0 to 1 · notable: a contested point in a live debate (also the default before judging). Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution
The complexity of tasks AI systems can complete autonomously has been growing rapidly year over year.
UnassessedNo current assessment. Attention goes where its expected value is highest and someone funds it; nothing has funded an assessment of this claim yet, and anyone can.constitutionempirical · derivedA factual claim that rests on inference from other evidence rather than direct observation.constitutionLLM capability trendsimportance · notableImportance 0.50, from 0 to 1 · notable: a contested point in a live debate (also the default before judging). Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution
Growth in spending on the largest AI training runs will slow substantially during the 2020s.
ContestedCredible evidence or argument exists on multiple sides.constitutionempirical · derivedA factual claim that rests on inference from other evidence rather than direct observation.constitutionAI Training Compute Cost Trendsimportance · notableImportance 0.55, from 0 to 1 · notable: a contested point in a live debate (also the default before judging). Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution
AI training runs costing on the order of $100 billion will be financially and technically feasible by 2030
UnassessedNo current assessment. Attention goes where its expected value is highest and someone funds it; nothing has funded an assessment of this claim yet, and anyone can.constitutionempirical · derivedA factual claim that rests on inference from other evidence rather than direct observation.constitutionAI Training Compute Cost TrendsTraining compute trendsimportance · notableImportance 0.50, from 0 to 1 · notable: a contested point in a live debate (also the default before judging). Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution
Generative AI assistance substantially reduces task completion time while maintaining or improving quality
UnassessedNo current assessment. Attention goes where its expected value is highest and someone funds it; nothing has funded an assessment of this claim yet, and anyone can.constitutionempirical · derivedA factual claim that rests on inference from other evidence rather than direct observation.constitutionGenerative AI productivity effectsimportance · notableImportance 0.55, from 0 to 1 · notable: a contested point in a live debate (also the default before judging). Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution
Surveys of AI researchers give even odds that machines outperform humans at all tasks around mid-century
UnassessedNo current assessment. Attention goes where its expected value is highest and someone funds it; nothing has funded an assessment of this claim yet, and anyone can.constitutionempirical · derivedA factual claim that rests on inference from other evidence rather than direct observation.constitutionHuman-level artificial intelligenceAI researcher survey forecastsimportance · minorImportance 0.35, from 0 to 1 · minor: narrow or largely settled, cheap to get right. Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution
AI systems will eventually match or exceed human performance on nearly all cognitive tasks
UnassessedNo current assessment. Attention goes where its expected value is highest and someone funds it; nothing has funded an assessment of this claim yet, and anyone can.constitutionempirical · derivedA factual claim that rests on inference from other evidence rather than direct observation.constitutionHuman-level artificial intelligenceimportance · majorImportance 0.70, from 0 to 1 · major: real consequence within a domain, actively argued. Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution
Consultants using AI on tasks beyond AI's capability frontier performed worse than consultants working without AI.
UnassessedNo current assessment. Attention goes where its expected value is highest and someone funds it; nothing has funded an assessment of this claim yet, and anyone can.constitutionempirical · derivedA factual claim that rests on inference from other evidence rather than direct observation.constitutionGenerative AI productivity effectsAI use beyond capability frontierimportance · minorImportance 0.35, from 0 to 1 · minor: narrow or largely settled, cheap to get right. Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution
The most expensive AI training run in 2025 cost on the order of several hundred million dollars
UnassessedNo current assessment. Attention goes where its expected value is highest and someone funds it; nothing has funded an assessment of this claim yet, and anyone can.constitutionempirical · derivedA factual claim that rests on inference from other evidence rather than direct observation.constitutionTraining compute trendsAI Training Compute Cost Trendsimportance · minorImportance 0.30, from 0 to 1 · minor: narrow or largely settled, cheap to get right. Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution
The most expensive published machine learning training run as of mid-2022 cost about 3.2 million dollars in compute.
UnassessedNo current assessment. Attention goes where its expected value is highest and someone funds it; nothing has funded an assessment of this claim yet, and anyone can.constitutionempirical · derivedA factual claim that rests on inference from other evidence rather than direct observation.constitutionTraining compute trendsAI Training Compute Cost Trendsimportance · minorImportance 0.35, from 0 to 1 · minor: narrow or largely settled, cheap to get right. Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution
The most expensive ML training run in 2021 cost less than 45 million dollars in compute.
UnassessedNo current assessment. Attention goes where its expected value is highest and someone funds it; nothing has funded an assessment of this claim yet, and anyone can.constitutionempirical · derivedA factual claim that rests on inference from other evidence rather than direct observation.constitutionTraining compute trendsimportance · minorImportance 0.35, from 0 to 1 · minor: narrow or largely settled, cheap to get right. Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution
Since September 2015, training costs for the largest-compute ML systems grew about 0.2 orders of magnitude per year, slower than the overall long-term trend.
UnassessedNo current assessment. Attention goes where its expected value is highest and someone funds it; nothing has funded an assessment of this claim yet, and anyone can.constitutionempirical · derivedA factual claim that rests on inference from other evidence rather than direct observation.constitutionTraining compute trendsAI Training Compute Cost Trendsimportance · notableImportance 0.50, from 0 to 1 · notable: a contested point in a live debate (also the default before judging). Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution
The dollar cost of compute for final ML training runs grew about 0.5 orders of magnitude per year from 2009 to 2022.
UnassessedNo current assessment. Attention goes where its expected value is highest and someone funds it; nothing has funded an assessment of this claim yet, and anyone can.constitutionempirical · derivedA factual claim that rests on inference from other evidence rather than direct observation.constitutionTraining compute trendsimportance · notableImportance 0.50, from 0 to 1 · notable: a contested point in a live debate (also the default before judging). Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution
Generative AI assistance in customer support work increases employee retention.
UnassessedNo current assessment. Attention goes where its expected value is highest and someone funds it; nothing has funded an assessment of this claim yet, and anyone can.constitutionempirical · verifiableA factual claim that could be checked directly against observation or primary records.constitutionGenerative AI and employmentAI earnings and labor market effectsimportance · minorImportance 0.30, from 0 to 1 · minor: narrow or largely settled, cheap to get right. Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution

Contribute

If a claim here is wrong, or missing evidence, open it: every claim page carries its own entry for challenges, evidence, and corrections. If the graph is missing a claim entirely, propose it below. A proposal is reviewed on its merits; accepted claims are matched against the graph and enter it with their reasoning on record.