[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-81788-en":3,"doc-seo-81788-105":29,"detail-sidebar-cat-0-en-105":90},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},81788,549758252649,"Ivy","https://ap-avatar.wpscdn.com/avatar/8000253669c5317157?_k=1778319167496531819",8,"Research & Report","Two AI Metrics Diverged Will it Make All the Difference","Exponential compute scaling raises a strategic question: do frontier AI models keep pulling away from what small-budget developers can access, or do capabilities converge as “meek models” expand? Building on prior theoretical work, the paper argues that the outcome hinges on how AI capability is valued and measured. While validation loss narrows its gap, other metrics let frontier models widen their lead indefinitely. The work provides mathematical conditions for when bounded metrics favor diffusion and when unbounded counterparts imply concentration.","Two AI Metrics Diverged: Will it Make All the Difference?  \nAlex Fogelson * 1 Zachary A. Brown * 1 Hans Gundlach 1 Jayson Lynch 1 Neil Thompson 1  \narXiv :2607 .009 13v 1 [ cs .AI] 1 Jul 2026  \nAbstract  \nAs exponential compute scaling continues, will the capabilities of frontier AI models outstrip what is accessible to developers on a small fixed budget? Or will capabilities converge, with “meek models inheriting the earth”? Building on Gundlach et al. (2025b), we show that the answer depends on how we value and measure AI capabilities. We discuss conventional performance measures and show that, while validation loss shows a shrinking gap, on other metrics frontier models grow their lead forever. Classifying performance metrics by their functional forms in relation to training (and inference) compute, we provide tight mathematical conditions for determining which metrics favor meek models, and show that bounded performance metrics always do. But careful interpretation of performance metrics is essential: we show that many common bounded metrics have closely-related counterpart metrics that are unbounded (and vice versa) . Determining the apt metric in a domain is a prerequisite for policy, since bounded and unbounded metrics may suggest opposing policy responses. If a particular capability — like software engineering, synthetic biology, or rhetorical persuasiveness—is unbounded when measured in the terms we care about, frontier-level capability will likely be concentrated in the hands of a few wealthy actors. Conversely, if that capability is  \ninstead bounded, frontier-level capabilities proliferate through meek models into the hands of the many.  \n*Equal contribution 1MIT FutureTech, CSAIL, Cambridge, MA, USA. Correspondence to: Alex Fogelson \u003Cfogel[son@mit.edu](son@mit.edu) > .  \nSecond Workshop on Technical AI Governance Research (TAIGR)@ ICML 2026, Seoul, South Korea. 2026. Copyright 2026 by the author(s) .  \n1. Introduction  \n1.1. Background  \nWhen AI models are trained with more compute they gain increased capabilities, as measured by (for example) validation loss (Kaplan et al., 2020 ; Rosenfeld, 2021 ; Hoffmann et al., 2022 ; Bahri et al., 2024) . Similarly, when AI models are allowed to reason about problems for longer—by using more compute at inference time—capabilities improve, as measured by (for example) success on benchmark tasks (Jones, 2021 ; Villalobos & Atkinson, 2023a) . As a result of these regularities, companies have invested exponentially increasing amounts in compute for pre-training and inference (Sevilla et al., 2022 ; Juniewicz, 2026) .  \nThese large expenditures have enabled a small handful of companies to offer more powerful AI capabilities than available elsewhere.  \nBut is this oligopolistic equilibrium guaranteed? Providers of smaller, cheaper open-weight models have been able to replicate the capabilities of expensive proprietary models with only a short delay—often less than a year (Cottier et al., 2024 ; Emberson, 2025) . If the capabilities gap doesn’t widen, but instead shrinks, we might expect frontier-level capabilities to diffuse very widely. In such a world, regulating AI capabilities may require compliance from many actors, or require enforcing regulations at the hardware level—which would require substantial new technological means and political will (O’Gara et al., 2025) .  \nGiven the regulatory relevance, it would be valuable to know whether the capabilities gap between expensive proprietary models and cheaper open models will shrink or widen. However the empirical literature is sparse, ambivalent, and—as we’ll see—sensitive to one’s individual utility function over a range of capabilities (Emberson, 2025 ; AI Index Steering Committee, 2025 ; Ihle, 2025) .  \nAgainst this backdrop, a recent paper Gundlach et al.(2025b) provides a theoretical argument that the gap could shrink. The authors (also authors of this paper) compare the performance of models with exponentially growing inve","cbCaiqswnv0Yehd9","https://ap.wps.com/l/cbCaiqswnv0Yehd9","pdf",605269,1,17,"English","en",105,"# Abstract\n# Introduction\n## Background\n## Capabilities Versus Metrics","[{\"question\":\"What question does the paper address about AI scaling?\",\"answer\":\"It asks whether frontier AI capabilities will outstrip what small-budget developers can access, or whether capabilities will converge as cheaper “meek” models improve.\"},{\"question\":\"How does the paper relate outcomes to performance metrics?\",\"answer\":\"It shows the answer depends on how capabilities are measured, because different metrics have different functional relationships to training and inference compute.\"},{\"question\":\"Why do bounded and unbounded metrics matter for policy?\",\"answer\":\"The paper argues that choosing a metric can lead to opposite policy implications: a bounded metric and its unbounded counterpart may suggest different responses regarding concentration versus diffusion of capabilities.\"}]",1784176147,43,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":85,"head_meta":87,"extra_data":89,"updated_unix":27},"two-ai-metrics-diverged-will-it-make-all-the-difference","",{"@graph":35,"@context":84},[36,53,67],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/two-ai-metrics-diverged-will-it-make-all-the-difference/81788/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":61,"encodingFormat":60,"isAccessibleForFree":62,"interactionStatistic":63},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-16",true,{"@type":64,"interactionType":65,"userInteractionCount":4},"InteractionCounter",{"@type":66},"ViewAction",{"@type":68,"mainEntity":69},"FAQPage",[70,76,80],{"name":71,"@type":72,"acceptedAnswer":73},"What question does the paper address about AI scaling?","Question",{"text":74,"@type":75},"It asks whether frontier AI capabilities will outstrip what small-budget developers can access, or whether capabilities will converge as cheaper “meek” models improve.","Answer",{"name":77,"@type":72,"acceptedAnswer":78},"How does the paper relate outcomes to performance metrics?",{"text":79,"@type":75},"It shows the answer depends on how capabilities are measured, because different metrics have different functional relationships to training and inference compute.",{"name":81,"@type":72,"acceptedAnswer":82},"Why do bounded and unbounded metrics matter for policy?",{"text":83,"@type":75},"The paper argues that choosing a metric can lead to opposite policy implications: a bounded metric and its unbounded counterpart may suggest different responses regarding concentration versus diffusion of capabilities.","https://schema.org",{"og:url":51,"og:type":86,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":88,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":91},[92,96,100,104,109,114,119,122,127,130,134],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":93,"show_sort_weight":94,"slug":95},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":97,"show_sort_weight":98,"slug":99},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":101,"show_sort_weight":102,"slug":103},"Exam",70,"exam",{"id":105,"doc_module":4,"doc_module_name":45,"category_name":106,"show_sort_weight":107,"slug":108},5,"Comic",60,"comic",{"id":110,"doc_module":4,"doc_module_name":45,"category_name":111,"show_sort_weight":112,"slug":113},6,"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":45,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":45,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":45,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":45,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":45,"category_name":136,"show_sort_weight":105,"slug":137},19,"General","general"]