{"id":105,"date":"2026-10-17T10:52:02","date_gmt":"2026-10-17T10:52:02","guid":{"rendered":"https:\/\/therisk.global\/ai-autonomous-systems\/?page_id=105"},"modified":"2026-10-21T14:00:33","modified_gmt":"2026-10-21T14:00:33","slug":"model-evaluation","status":"publish","type":"page","link":"https:\/\/therisk.global\/ai-autonomous-systems\/topics\/model-evaluation\/","title":{"rendered":"Model evaluation"},"content":{"rendered":"\n<!-- Nexus network page \u00b7 https:\/\/therisk.global\/ai-autonomous-systems\/topics\/model-evaluation\/ \u00b7 source authored \u00b7 generated by integrations\/wordpress\/network\/build.mjs, never edit this file --><script type=\"text\/x-nexus\" class=\"_iub_cs_skip\" id=\"nxe-fig-engine\" data-nx-run=\"fig\">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<\/script><script type=\"application\/json\" class=\"_iub_cs_skip\" id=\"nxe-water-data\" data-nx-pack=\"deflate\">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<\/script><script type=\"application\/json\" class=\"_iub_cs_skip\" id=\"nxe-water-gibs\" data-nx-pack=\"deflate\">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<\/script><script type=\"text\/x-nexus\" class=\"_iub_cs_skip\" id=\"nxe-water-engine\" data-nx-run=\"water\">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<\/script><script type=\"text\/x-nexus\" class=\"_iub_cs_skip\" id=\"nxe-maplibre-engine\" data-nx-run=\"maplibre\">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<\/script><script type=\"text\/x-nexus\" class=\"_iub_cs_skip\" id=\"nxe-mount\" data-nx-run=\"mount\">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<\/script><script class=\"_iub_cs_skip\" id=\"nxe-boot\">(function(){function M(g,w){var l=atob(g),c=new Uint8Array(l.length),r;for(r=0;r< l.length;r++){c[r]=l.charCodeAt(r)}var o=0,e=0,b=0,f=new Uint8Array(Math.max(1024,c.length*6)),n=0;var V=[3,4,5,6,7,8,9,10,11,13,15,17,19,23,27,31,35,43,51,59,67,83,99,115,131,163,195,227,258];var W=[0,0,0,0,0,0,0,0,1,1,1,1,2,2,2,2,3,3,3,3,4,4,4,4,5,5,5,5,0];var Y=[1,2,3,4,5,7,9,13,17,25,33,49,65,97,129,193,257,385,513,769,1025,1537,2049,3073,4097,6145,8193,12289,16385,24577];var Z=[0,0,0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10,11,11,12,12,13,13];var z=[16,17,18,0,8,7,9,6,10,5,11,4,12,3,13,2,14,1,15];function u(a){while(b< a){if(o>=c.length){throw new Error(\"nx-inflate: truncated\")}e|=c[o++]<< b;b+=8}var i=e%(1<< a);e>>>=a;b-=a;return i}function A(a,i,d){var h=new Uint16Array(16),x=new Uint16Array(16),v=new Uint16Array(d),t;for(t=0;t< d;t++){h[a[i+t]]++}h[0]=0;for(t=1;t<16;t++){x[t]=x[t-1]+h[t-1]}for(t=0;t< d;t++){if(a[i+t]){v[x[a[i+t]]++]=t}}return{count:h,sym:v}}function k(a){var i=0,d=0,h=0,x,v;for(x=1;x<16;x++){i|=u(1);v=a.count[x];if(i-v< d){return a.sym[h+(i-d)]}h+=v;d+=v;d<<=1;i<<=1}throw new Error(\"nx-inflate: bad code\")}function I(a){if(n+a<=f.length){return}var i=new Uint8Array(Math.max(f.length*2,n+a));i.set(f.subarray(0,n));f=i}var L=null,R=null,T=0;while(!T){T=u(1);var O=u(2),S,$;if(O===0){e=0;b=0;var E=c[o]|c[o+1]<<8;o+=4;I(E);f.set(c.subarray(o,o+E),n);n+=E;o+=E;continue}if(O===1){if(!L){var X=new Uint8Array(288),J=new Uint8Array(30);for(r=0;r<288;r++){X[r]=r<144?8:r<256?9:r<280?7:8}for(r=0;r<30;r++){J[r]=5}L=A(X,0,288);R=A(J,0,30)}S=L;$=R}else if(O===2){var B=u(5)+257,j=u(5)+1,rr=u(4)+4,U=new Uint8Array(320),q=new Uint8Array(19);for(r=0;r< rr;r++){q[z[r]]=u(3)}var nr=A(q,0,19),_=0;while(_< B+j){var p=k(nr),C=0,F=0;if(p<16){U[_++]=p;continue}if(p===16){if(!_){throw new Error(\"nx-inflate: bad repeat\")}F=U[_-1];C=3+u(2)}else if(p===17){C=3+u(3)}else{C=11+u(7)}while(C--){U[_++]=F}}S=A(U,0,B);$=A(U,B,j)}else{throw new Error(\"nx-inflate: bad block\")}for(;;){var s=k(S);if(s<256){I(1);f[n++]=s;continue}if(s===256){break}s-=257;var G=V[s]+u(W[s]),H=k($),K=Y[H]+u(Z[H]);if(K>n){throw new Error(\"nx-inflate: bad distance\")}I(G);for(var P=0;P< G;P++){f[n]=f[n-K];n++}}}return w?f.subarray(0,n):new TextDecoder(\"utf-8\").decode(f.subarray(0,n))}if(typeof window.__nxInflate!==\"function\"){window.__nxInflate=M}function Q(g){var w=M(g,true),l=w[0]|w[1]<<8|w[2]<<16|w[3]<<24;var c=w.subarray(4+l);function r(o,e){var b=\"\",f=c.subarray(o,o+e);for(var n=0;e>n;n+=8192){b+=String.fromCharCode.apply(null,f.subarray(n,n+8192))}return btoa(b)}return JSON.parse(new TextDecoder(\"utf-8\").decode(w.subarray(4,4+l)),function(o,e){return e?e.$b?r(e.$b[0],e.$b[1]):e:e})}if(typeof window.__nxUnpack!==\"function\"){window.__nxUnpack=Q}var N=document.querySelectorAll(\"script[data-nx-run]\");for(var D=0;D< N.length;D++){var y=N[D];if(y.hasAttribute(\"data-nx-ran\")){continue}y.setAttribute(\"data-nx-ran\",\"true\");try{var m=document.createElement(\"script\");m.className=\"_iub_cs_skip\";m.setAttribute(\"data-nx-runtime\",y.getAttribute(\"data-nx-run\"));m.textContent=window.__nxInflate(y.textContent);(document.head||document.documentElement).appendChild(m)}catch(g){y.setAttribute(\"data-nx-ran\",\"failed\")}}})();<\/script><div class=\"nxe-host\" data-nxe-embed=\"n-g-p08-topics-model-evaluation\"><div class=\"nxe\" id=\"nxe-n5fmk9bk\" data-inst=\"gcri\" data-theme=\"light\" data-grammar=\"atlas\" data-domain=\"project\" data-arche=\"topic\" data-density=\"compact\" lang=\"en\" data-nx-css=\"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\"><style>#nxe-n5fmk9bk.nxe{display:block;box-sizing:border-box;max-width:1180px;margin:0 auto;padding:0 16px;font-family:ui-sans-serif,system-ui,-apple-system,\"Segoe UI\",sans-serif;line-height:1.55;overflow-wrap:break-word}#nxe-n5fmk9bk.nxe :where(*,*::before,*::after){box-sizing:inherit}#nxe-n5fmk9bk.nxe :where(section,header,footer,nav,aside){display:block;margin:0 0 32px}#nxe-n5fmk9bk.nxe :where(h1,h2,h3,h4){line-height:1.2;margin:24px 0 12px;font-weight:600}#nxe-n5fmk9bk.nxe :where(p,ul,ol,dl,table,figure,details){margin:0 0 16px}#nxe-n5fmk9bk.nxe :where(ul,ol){padding-left:20px}#nxe-n5fmk9bk.nxe a{color:inherit;text-decoration:underline;text-underline-offset:2px}#nxe-n5fmk9bk.nxe table{border-collapse:collapse;width:100%;display:block;overflow-x:auto}#nxe-n5fmk9bk.nxe :where(th,td){padding:6px 8px;text-align:left;vertical-align:top;border-bottom:1px solid;border-color:inherit}#nxe-n5fmk9bk.nxe svg{display:none}#nxe-n5fmk9bk.nxe svg.nx-i{display:inline-block;width:18px;height:18px;vertical-align:-3px}#nxe-n5fmk9bk.nxe :where([role=\"tablist\"],[role=\"toolbar\"],[hidden],button,.nx-theme-toggle,.nx-live-slot){display:none}#nxe-n5fmk9bk.nxe img{max-width:100%;height:auto}<\/style><svg class=\"nx-defs\" aria-hidden=\"true\" focusable=\"false\" width=\"0\" height=\"0\"><defs><linearGradient id=\"nx-gi\" x1=\"0\" y1=\"0\" x2=\"1\" y2=\"1\"><stop offset=\"0\" class=\"nx-gi-a\"\/><stop offset=\"0.55\" class=\"nx-gi-b\"\/><stop offset=\"1\" class=\"nx-gi-c\"\/><\/linearGradient><linearGradient id=\"nx-giu\" gradientUnits=\"userSpaceOnUse\" x1=\"0\" y1=\"0\" x2=\"800\" y2=\"0\"><stop offset=\"0\" class=\"nx-gi-a\"\/><stop offset=\"0.55\" class=\"nx-gi-b\"\/><stop offset=\"1\" class=\"nx-gi-c\"\/><\/linearGradient><linearGradient id=\"nx-gp\" x1=\"0\" y1=\"0\" x2=\"1\" y2=\"1\"><stop offset=\"0\" class=\"nx-gp-a\"\/><stop offset=\"1\" class=\"nx-gp-b\"\/><\/linearGradient><\/defs><\/svg><section class=\"nx-hero\" data-layout=\"editorial\" aria-labelledby=\"nx-title-1\"><div class=\"nx-container\"><div class=\"nx-theme-toggle\" role=\"group\" aria-label=\"Appearance of this page\" hidden><button type=\"button\" data-mode=\"light\" aria-pressed=\"false\">Light<\/button><button type=\"button\" data-mode=\"dark\" aria-pressed=\"false\">Dark<\/button><button type=\"button\" data-mode=\"auto\" aria-pressed=\"true\">Auto<\/button><\/div><div class=\"nx-folio\" aria-label=\"Document reference\"><strong>G-P08 \u00b7 TOPIC \u00b7 model-evaluation<\/strong>GCRI dossier \u00b7 service level R0<\/div><div class=\"nx-hero-grid\"><div class=\"nx-hero-text\"><div class=\"nx-eyebrow-row\"><span class=\"nx-eyebrow\">GCRI<\/span><span class=\"nx-eyebrow\" aria-hidden=\"true\">\u00b7<\/span><span class=\"nx-eyebrow\">Artificial Intelligence and Autonomous Systems<\/span><span class=\"nx-eyebrow\" aria-hidden=\"true\">\u00b7<\/span><span class=\"nx-eyebrow\">In development<\/span><\/div><div class=\"nx-accent-row\"><span class=\"nx-accent-rule\" aria-hidden=\"true\"><\/span><span>Topic dossier \u00b7 Artificial Intelligence and Autonomous Systems<\/span><\/div><h1 id=\"nx-title-1\" class=\"nx-hero-title\">Model evaluation<\/h1><p class=\"nx-tagline\">Model evaluation asks what an AI model can do, how reliably, and where it fails. Benchmarks, red teaming, human studies and field monitoring each measure something different, and none fully predicts behaviour in deployment. This page sets out how evaluations are designed and where they mislead, how results should be reported, and what evidence serves the developers, deployers and regulators who rely on them.<\/p><div class=\"nx-actions\"><a class=\"nx-btn\" data-variant=\"primary\" href=\"https:\/\/therisk.global\/ai-autonomous-systems\/contact\/\">Make an enquiry <svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\" fill=\"none\" stroke=\"currentColor\" stroke-width=\"1.5\" stroke-linecap=\"round\" stroke-linejoin=\"round\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><a class=\"nx-btn\" data-variant=\"secondary\" href=\"https:\/\/therisk.global\/ai-autonomous-systems\/topics\/\">Back to Topics<\/a><\/div><\/div><div class=\"nx-hero-panel\"><div class=\"nx-register\" data-variant=\"contents\"><div class=\"nx-register-head\"><span class=\"nx-eyebrow\"><strong>In this dossier<\/strong><\/span><span class=\"nx-eyebrow\">7 parts<\/span><\/div><ol class=\"nx-register-rows\"><li><a href=\"#nx-sec-brief-answer\"><span class=\"nx-register-n\">01<\/span><span class=\"nx-register-k\">In brief<\/span><span class=\"nx-register-v\">The short answer<\/span><\/a><\/li><li><a href=\"#nx-sec-question\"><span class=\"nx-register-n\">02<\/span><span class=\"nx-register-k\">The question<\/span><span class=\"nx-register-v\">Model evaluation: how the risk develops<\/span><\/a><\/li><li><a href=\"#nx-sec-terms\"><span class=\"nx-register-n\">03<\/span><span class=\"nx-register-k\">Key terms<\/span><span class=\"nx-register-v\">Terms used on this page<\/span><\/a><\/li><li><a href=\"#nx-sec-indicators\"><span class=\"nx-register-n\">04<\/span><span class=\"nx-register-k\">Indicators<\/span><span class=\"nx-register-v\">What is measured, and how well<\/span><\/a><\/li><li><a href=\"#nx-sec-pipeline\"><span class=\"nx-register-n\">05<\/span><span class=\"nx-register-k\">Canvas<\/span><span class=\"nx-register-v\">From sources to decisions<\/span><\/a><\/li><li><a href=\"#nx-sec-sources\"><span class=\"nx-register-n\">06<\/span><span class=\"nx-register-k\">Sources<\/span><span class=\"nx-register-v\">Where the evidence comes from<\/span><\/a><\/li><li><a href=\"#nx-sec-decisions\"><span class=\"nx-register-n\">07<\/span><span class=\"nx-register-k\">Decisions<\/span><span class=\"nx-register-v\">Who decides, and what the evidence serves<\/span><\/a><\/li><\/ol><p class=\"nx-register-foot\">GCRI \u00b7 In development \u00b7 service level R0<\/p><\/div><\/div><\/div><\/div><\/section> <nav class=\"nx-index-strip\" data-nx-spy=\"true\" aria-label=\"On this page\"><div class=\"nx-container\"><a href=\"#nx-sec-brief-answer\"><span>01<\/span>The short answer<\/a><a href=\"#nx-sec-field\"><span>02<\/span>Connected activity in East Asia<\/a><a href=\"#nx-s-4-sec\"><span>03<\/span>Model registry, in outline<\/a><a href=\"#nx-sec-question\"><span>04<\/span>Model evaluation: how the risk\u2026<\/a><a href=\"#nx-sec-terms\"><span>05<\/span>Terms used on this page<\/a><a href=\"#nx-sec-indicators\"><span>06<\/span>What is measured, and how well<\/a><a href=\"#nx-sec-pipeline\"><span>07<\/span>From sources to decisions<\/a><a href=\"#nx-sec-sources\"><span>08<\/span>Where the evidence comes from<\/a><\/div><\/nav> <section id=\"nx-sec-brief-answer\" class=\"nx-section\" data-hue=\"project\" data-variant=\"stack\" aria-labelledby=\"nx-s-2\"><div class=\"nx-container\"><div class=\"nx-section-head\"><div class=\"nx-eyebrow-row\"><span class=\"nx-eyebrow\">01<\/span><span class=\"nx-rule\" aria-hidden=\"true\"><\/span><span class=\"nx-eyebrow\">In brief<\/span><\/div><h2 id=\"nx-s-2\" class=\"nx-section-title\">The short answer.<\/h2><p class=\"nx-lede\">AI model evaluation is the measurement of what a model can do, how reliably and where it fails. Benchmarks, red teaming, human studies and monitoring after deployment each measure different things, and none fully predicts real use, so results carry uncertainty, contamination checks and stated limits.<\/p><\/div><\/div><\/section><section id=\"nx-sec-field\" class=\"nx-section\" data-hue=\"infrastructure\" data-tone=\"alt\" aria-labelledby=\"nx-s-3\"><div class=\"nx-container\"><div class=\"nx-section-head\"><div class=\"nx-eyebrow-row\"><span class=\"nx-eyebrow\">02<\/span><span class=\"nx-rule\" aria-hidden=\"true\"><\/span><span class=\"nx-eyebrow\">Geography<\/span><\/div><h2 id=\"nx-s-3\" class=\"nx-section-title\">Connected activity in East Asia.<\/h2><p class=\"nx-lede\">Night-time lights and population density, the footprint of electrified, connected activity, framed on East Asia; each layer is a dated NASA product.<\/p><\/div><div class=\"nx-section-body\"><div class=\"nx-canvas-wrap\"><div class=\"nx-cmap-field\" data-nx-map=\"ctx-field\" data-cmap=\"east-asia:digital\"><div class=\"nx-wm\" data-nx-water=\"ctx-digital\" data-wm-domain=\"ctx-digital\" data-wm-tiles=\"on\" data-wm-flat=\"on\" data-wm-focus=\"100,18,146,50\" data-ws-live=\"off\" data-claim=\"INTEGRATED\" data-wm-panel=\"closed\" data-wm-mode=\"2d\"><div class=\"nx-wm-stage\" tabindex=\"0\" role=\"application\" aria-roledescription=\"interactive map\" aria-label=\"Connected activity, East Asia: satellite and model layers from NASA on a 3D terrain of ASTER GDEM elevation. Arrow keys move; with Shift they turn and tilt in 3D; plus and minus zoom, 0 resets; I inspects, question mark lists the keys.\" aria-describedby=\"nx-wm-help\"><canvas class=\"nx-wm-canvas\" aria-hidden=\"true\"><\/canvas><p class=\"nx-wm-noscript\">The interactive map needs JavaScript; the layer list and sources are below.<\/p><div class=\"nx-wm-presets\" role=\"group\" aria-label=\"Quick views\"><button type=\"button\" class=\"nx-wm-preset\" data-wm-preset=\"ntl||||\" aria-pressed=\"false\">Night lights<\/button><button type=\"button\" class=\"nx-wm-preset\" data-wm-preset=\"people||||\" aria-pressed=\"false\">Population<\/button><\/div><div class=\"nx-wm-panel\" id=\"nx-wm-panel\" role=\"region\" aria-label=\"Map layers\"><div class=\"nx-wm-panel-head\"><p class=\"nx-wm-panel-title\">Layers <span class=\"nx-wm-count\">2<\/span><\/p><button type=\"button\" class=\"nx-wm-icon nx-wm-panel-close\" data-wm-act=\"panel\" aria-label=\"Close the layer list\"><svg viewBox=\"0 0 16 16\" aria-hidden=\"true\" focusable=\"false\"><path d=\"M3.5 3.5l9 9M12.5 3.5l-9 9\" fill=\"none\" stroke=\"currentColor\" stroke-width=\"1.5\" stroke-linecap=\"round\" stroke-linejoin=\"round\"\/><\/svg><\/button><\/div><div class=\"nx-wm-panel-body\"><div class=\"nx-wm-group\"><p class=\"nx-wm-group-title\">Connected activity<\/p><button type=\"button\" class=\"nx-wm-layer\" data-wm-toggle=\"ntl\" aria-pressed=\"true\" data-wm-z=\"20\" data-wm-lag=\"0\" data-wm-op=\"0.9\" data-wm-name=\"Night-time lights, daily\" data-wm-short=\"Night lights\" data-wm-src=\"NASA LAADS DAAC VIIRS Black Marble (VNP46A2)\" data-wm-prov=\"NASA LAADS DAAC\" data-wm-kind=\"Satellite radiance, corrected for moonlight and atmosphere\"><span class=\"nx-wm-check\" aria-hidden=\"true\"><\/span><span class=\"nx-wm-layer-text\"><span class=\"nx-wm-layer-name\">Night-time lights, daily<\/span><span class=\"nx-wm-layer-src\">NASA LAADS DAAC VIIRS Black Marble (VNP46A2) \u00b7 <span data-wm-date=\"ntl\">12 Oct 2026<\/span><\/span><\/span><\/button><div class=\"nx-wm-tools\" data-wm-tools=\"ntl\" hidden><\/div><button type=\"button\" class=\"nx-wm-layer\" data-wm-toggle=\"people\" aria-pressed=\"true\" data-wm-z=\"21\" data-wm-lag=\"0\" data-wm-op=\"0.8\" data-wm-name=\"Population density, 2020\" data-wm-short=\"Population density\" data-wm-src=\"NASA SEDAC (CIESIN, Columbia University) Gridded Population of the World v4.11\" data-wm-prov=\"NASA SEDAC (CIESIN, Columbia University)\" data-wm-kind=\"Census-based estimate, static (2020)\"><span class=\"nx-wm-check\" aria-hidden=\"true\"><\/span><span class=\"nx-wm-layer-text\"><span class=\"nx-wm-layer-name\">Population density, 2020<\/span><span class=\"nx-wm-layer-src\">NASA SEDAC (CIESIN, Columbia University) Gridded Population of the World v4.11 \u00b7 <span data-wm-date=\"people\">static<\/span><\/span><\/span><\/button><div class=\"nx-wm-tools\" data-wm-tools=\"people\" hidden><\/div><button type=\"button\" class=\"nx-wm-layer\" data-wm-toggle=\"true\" aria-pressed=\"false\" data-wm-z=\"22\" data-wm-lag=\"1\" data-wm-name=\"True colour\" data-wm-short=\"True colour\" data-wm-src=\"NASA LANCE VIIRS NOAA-20 corrected reflectance\" data-wm-prov=\"NASA LANCE\" data-wm-kind=\"Satellite image\"><span class=\"nx-wm-check\" aria-hidden=\"true\"><\/span><span class=\"nx-wm-layer-text\"><span class=\"nx-wm-layer-name\">True colour<\/span><span class=\"nx-wm-layer-src\">NASA LANCE VIIRS NOAA-20 corrected reflectance \u00b7 <span data-wm-date=\"true\">13 Oct 2026<\/span><\/span><\/span><\/button><div class=\"nx-wm-tools\" data-wm-tools=\"true\" hidden><\/div><\/div><\/div><\/div><div class=\"nx-wm-rail\" role=\"toolbar\" aria-orientation=\"vertical\" aria-label=\"Map controls\"><button type=\"button\" class=\"nx-wm-icon nx-wm-open\" data-wm-act=\"panel\" aria-controls=\"nx-wm-panel\" aria-expanded=\"false\" aria-label=\"Layers\" title=\"Layers\"><svg viewBox=\"0 0 16 16\" aria-hidden=\"true\" focusable=\"false\"><path d=\"M8 2.5l-6.5 3.5 6.5 3.5 6.5-3.5zm-6.5 6.5 6.5 3.5 6.5-3.5m-13 2.8 6.5 3.5 6.5-3.5\" fill=\"none\" stroke=\"currentColor\" stroke-width=\"1.5\" stroke-linecap=\"round\" stroke-linejoin=\"round\"\/><\/svg><\/button><button type=\"button\" class=\"nx-wm-icon\" data-wm-act=\"tools\" aria-controls=\"nx-wm-menu\" aria-expanded=\"false\" aria-label=\"Expert tools\" title=\"Expert tools\"><svg viewBox=\"0 0 16 16\" aria-hidden=\"true\" focusable=\"false\"><path d=\"M3 4.5l10 0m-10 3.5 10 0m-10 3.5 10 0m-7.5-8.5 0 3m5 .5 0 3m-3.5.5 0 3\" fill=\"none\" stroke=\"currentColor\" stroke-width=\"1.5\" stroke-linecap=\"round\" stroke-linejoin=\"round\"\/><\/svg><\/button><span class=\"nx-wm-sep\" aria-hidden=\"true\"><\/span><button type=\"button\" class=\"nx-wm-icon\" data-wm-act=\"in\" aria-label=\"Zoom in\" title=\"Zoom in\"><svg viewBox=\"0 0 16 16\" aria-hidden=\"true\" focusable=\"false\"><path d=\"M8 3v10M3 8h10\" fill=\"none\" stroke=\"currentColor\" stroke-width=\"1.5\" stroke-linecap=\"round\" stroke-linejoin=\"round\"\/><\/svg><\/button><button type=\"button\" class=\"nx-wm-icon\" data-wm-act=\"out\" aria-label=\"Zoom out\" title=\"Zoom out\"><svg viewBox=\"0 0 16 16\" aria-hidden=\"true\" focusable=\"false\"><path d=\"M3 8h10\" fill=\"none\" stroke=\"currentColor\" stroke-width=\"1.5\" stroke-linecap=\"round\" stroke-linejoin=\"round\"\/><\/svg><\/button><button type=\"button\" class=\"nx-wm-icon\" data-wm-act=\"reset\" aria-label=\"Reset the view\" title=\"Reset the view\"><svg viewBox=\"0 0 16 16\" aria-hidden=\"true\" focusable=\"false\"><path d=\"M3.2 8a4.8 4.8 0 1 0 1.5-3.5M3.2 2.6v2.6h2.6\" fill=\"none\" stroke=\"currentColor\" stroke-width=\"1.5\" stroke-linecap=\"round\" stroke-linejoin=\"round\"\/><\/svg><\/button><button type=\"button\" class=\"nx-wm-icon\" data-wm-act=\"full\" aria-label=\"Full screen\" title=\"Full screen\"><svg viewBox=\"0 0 16 16\" aria-hidden=\"true\" focusable=\"false\"><path d=\"M2.5 6l0-3.5 3.5 0m7.5 3.5 0-3.5-3.5 0m-7.5 7.5 0 3.5 3.5 0m7.5-3.5 0 3.5-3.5 0\" fill=\"none\" stroke=\"currentColor\" stroke-width=\"1.5\" stroke-linecap=\"round\" stroke-linejoin=\"round\"\/><\/svg><\/button><\/div><div class=\"nx-wm-menu\" id=\"nx-wm-menu\" role=\"menu\" aria-label=\"Expert tools\" hidden><button type=\"button\" class=\"nx-wm-menu-item\" role=\"menuitemcheckbox\" aria-checked=\"false\" data-wm-tool=\"inspect\">Inspect a point<kbd>I<\/kbd><\/button><button type=\"button\" class=\"nx-wm-menu-item\" role=\"menuitemcheckbox\" aria-checked=\"false\" data-wm-tool=\"measure\">Measure distance<kbd>M<\/kbd><\/button><button type=\"button\" class=\"nx-wm-menu-item\" role=\"menuitemcheckbox\" aria-checked=\"false\" data-wm-tool=\"compare\">Compare two dates<kbd>C<\/kbd><\/button><button type=\"button\" class=\"nx-wm-menu-item\" role=\"menuitem\" data-wm-tool=\"share\">Copy a link to this view<\/button><button type=\"button\" class=\"nx-wm-menu-item\" role=\"menuitem\" data-wm-tool=\"png\">Export as PNG with credits<\/button><button type=\"button\" class=\"nx-wm-menu-item\" role=\"menuitem\" data-wm-tool=\"keys\">Keyboard shortcuts<kbd>?<\/kbd><\/button><\/div><div class=\"nx-wm-keys\" role=\"dialog\" aria-label=\"Keyboard shortcuts\" hidden><p class=\"nx-wm-keys-title\">Keyboard, with the map focused<\/p><dl class=\"nx-wm-keys-list\"><div><dt><kbd>Arrows<\/kbd><\/dt><dd>Move<\/dd><\/div><div><dt><kbd>Shift and arrows<\/kbd><\/dt><dd>Turn and tilt (3D)<\/dd><\/div><div><dt><kbd>+ and \u2212<\/kbd><\/dt><dd>Zoom<\/dd><\/div><div><dt><kbd>0<\/kbd><\/dt><dd>Reset the view<\/dd><\/div><div><dt><kbd>N and P<\/kbd><\/dt><dd>Next and previous station<\/dd><\/div><div><dt><kbd>Enter<\/kbd><\/dt><dd>Pin the station<\/dd><\/div><div><dt><kbd>I<\/kbd><\/dt><dd>Inspect a point<\/dd><\/div><div><dt><kbd>M<\/kbd><\/dt><dd>Measure<\/dd><\/div><div><dt><kbd>C<\/kbd><\/dt><dd>Compare dates<\/dd><\/div><div><dt><kbd>Escape<\/kbd><\/dt><dd>Close<\/dd><\/div><\/dl><button type=\"button\" class=\"nx-wm-icon nx-wm-keys-close\" data-wm-tool=\"keys\" aria-label=\"Close the shortcuts\"><svg viewBox=\"0 0 16 16\" aria-hidden=\"true\" focusable=\"false\"><path d=\"M3.5 3.5l9 9M12.5 3.5l-9 9\" fill=\"none\" stroke=\"currentColor\" stroke-width=\"1.5\" stroke-linecap=\"round\" stroke-linejoin=\"round\"\/><\/svg><\/button><\/div><p class=\"nx-wm-hint\" hidden>Hold Ctrl or \u2318 and scroll to zoom<\/p><p class=\"nx-wm-toolbar\" role=\"status\" hidden><\/p><div class=\"nx-wm-tip\" role=\"tooltip\" hidden><\/div><div class=\"nx-wm-pin\" hidden><div class=\"nx-wm-pin-head\"><p class=\"nx-wm-pin-title\"><\/p><button type=\"button\" class=\"nx-wm-icon\" data-wm-act=\"unpin\" aria-label=\"Close the card\"><svg viewBox=\"0 0 16 16\" aria-hidden=\"true\" focusable=\"false\"><path d=\"M3.5 3.5l9 9M12.5 3.5l-9 9\" fill=\"none\" stroke=\"currentColor\" stroke-width=\"1.5\" stroke-linecap=\"round\" stroke-linejoin=\"round\"\/><\/svg><\/button><\/div><div class=\"nx-wm-pin-chart\"><\/div><div class=\"nx-wm-pin-body\"><\/div><\/div><div class=\"nx-wm-dock\"><div class=\"nx-wm-time\" role=\"group\" aria-label=\"Imagery date\"><button type=\"button\" class=\"nx-wm-icon\" data-wm-step=\"prev\" aria-label=\"One day earlier\">\u2039<\/button><p class=\"nx-wm-time-label\">Latest available<\/p><button type=\"button\" class=\"nx-wm-icon\" data-wm-step=\"next\" aria-label=\"One day later\">\u203a<\/button><span class=\"nx-wm-track\" role=\"slider\" tabindex=\"0\" aria-label=\"Imagery date, up to 30 days back\"><\/span><button type=\"button\" class=\"nx-wm-latest\" data-wm-step=\"latest\" aria-pressed=\"true\">Latest<\/button><\/div><div class=\"nx-wm-chips\" role=\"group\" aria-label=\"Legends\"><\/div><p class=\"nx-wm-attrib\">NASA GIBS \/ EOSDIS \u00b7 ASTER GDEM (METI, NASA) \u00b7 Natural Earth<\/p><div class=\"nx-wm-legends\" aria-live=\"polite\" hidden><\/div><\/div><\/div><div class=\"nx-container nx-wm-below\"><p class=\"nx-wm-status\" role=\"status\">Loading the latest NASA imagery.<\/p><p id=\"nx-wm-help\" class=\"nx-wm-help\">Drag to move. Ctrl or \u2318 and scroll, pinch or the buttons to zoom; hover for the values under the pointer; reset returns to the region.<\/p><p class=\"nx-caption\"><span>Night-time lights and population density, the footprint of electrified, connected activity, framed on East Asia.<\/span><span data-honesty=\"true\">Satellite and model products as NASA publishes them, each with its own date; indicators, not assessments of any place; claim ceiling INTEGRATED.<\/span><\/p><details class=\"nx-wm-credits\"><summary><span class=\"nx-wm-credits-line\">Data: NASA GIBS \u00b7 ASTER GDEM \u00b7 Natural Earth<\/span><span class=\"nx-wm-credits-more\">Sources and licences<\/span><\/summary><p class=\"nx-wm-credit\"><ul class=\"nx-wm-credit-list\"><li data-wm-credit=\"nasa.gibs.field-layers\">We acknowledge the use of imagery provided by services from NASA's Global Imagery Browse Services (GIBS), part of NASA's Earth Science Data and Information System (ESDIS). Colours redrawn from the published colour map; values averaged into H3 cells where shown (modified). <a class=\"nx-link\" href=\"https:\/\/www.earthdata.nasa.gov\/engage\/open-data-services-software-policies\" rel=\"noopener\" target=\"_blank\">NASA open data<\/a><\/li><li data-wm-credit=\"nasa.gibs.water-layers\">We acknowledge the use of imagery provided by services from NASA's Global Imagery Browse Services (GIBS), part of NASA's Earth Science Data and Information System (ESDIS). Colours redrawn from the published colour map; values averaged into H3 cells where shown (modified). <a class=\"nx-link\" href=\"https:\/\/www.earthdata.nasa.gov\/engage\/open-data-services-software-policies\" rel=\"noopener\" target=\"_blank\">NASA open data<\/a><\/li><li data-wm-credit=\"nasa.laads.black-marble\">VIIRS Black Marble night-time lights, NASA Goddard Space Flight Center; distributed by the NASA Level-1 and Atmosphere Archive and Distribution System (LAADS DAAC); imagery via NASA GIBS. Colours redrawn from the published colour map; values averaged into H3 cells where shown (modified). <a class=\"nx-link\" href=\"https:\/\/www.earthdata.nasa.gov\/engage\/open-data-services-software-policies\" rel=\"noopener\" target=\"_blank\">NASA open data<\/a><\/li><li data-wm-credit=\"sedac.gpw-v4\">Center for International Earth Science Information Network (CIESIN), Columbia University. 2018. Gridded Population of the World, Version 4 (GPWv4): Population Density, Revision 11. NASA Socioeconomic Data and Applications Center (SEDAC). Licensed under CC BY 4.0. Colours redrawn from the published colour map; values averaged into H3 cells where shown (modified). <a class=\"nx-link\" href=\"https:\/\/creativecommons.org\/licenses\/by\/4.0\/\" rel=\"noopener\" target=\"_blank\">CC BY 4.0<\/a><\/li><li data-wm-credit=\"nasa.lance.viirs-water\">VIIRS near-real-time products (flood, land surface temperature, snow cover, corrected reflectance), NASA LANCE; imagery via NASA GIBS. Colours redrawn from the published colour map; values averaged into H3 cells where shown (modified). <a class=\"nx-link\" href=\"https:\/\/www.earthdata.nasa.gov\/engage\/open-data-services-software-policies\" rel=\"noopener\" target=\"_blank\">NASA open data<\/a><\/li><li data-wm-credit=\"nasa-meti.aster-gdem-v3\">ASTER GDEM is a product of METI and NASA (ASTER Global Digital Elevation Model V003, NASA LP DAAC); imagery via NASA GIBS. Colours redrawn from the published colour map; values averaged into H3 cells where shown (modified). <a class=\"nx-link\" href=\"https:\/\/www.earthdata.nasa.gov\/engage\/open-data-services-software-policies\" rel=\"noopener\" target=\"_blank\">NASA open data<\/a><\/li><li data-wm-credit=\"naturalearth.all-scales\">Made with Natural Earth. Free vector and raster map data @ naturalearthdata.com. <a class=\"nx-link\" href=\"https:\/\/www.naturalearthdata.com\/about\/terms-of-use\/\" rel=\"noopener\" target=\"_blank\">Public domain<\/a><\/li><li>Globe: MapLibre GL JS 6.9.1, BSD-3-Clause, \u00a9 MapLibre contributors.<\/li><\/ul><p class=\"nx-wm-credit\"><strong>Layers and dates<\/strong><\/p><ul class=\"nx-wm-credit-list\"><li>Night-time lights, daily: VIIRS Black Marble (VNP46A2), NASA LAADS DAAC \u00b7 <span data-wm-date=\"ntl\">12 Oct 2026<\/span><\/li><li>Population density, 2020: Population: CIESIN, Columbia University, Gridded Population of the World v4.11 (2018), NASA SEDAC, CC BY 4.0 \u00b7 <span data-wm-date=\"people\">static<\/span><\/li><li>True colour: VIIRS NOAA-20 corrected reflectance, NASA LANCE \u00b7 <span data-wm-date=\"true\">13 Oct 2026<\/span><\/li><\/ul><\/p><\/details><\/div><\/div><div class=\"nx-live-slot nx-live-slot--map\" data-nxf-slot=\"map\" data-nxf-kind=\"map\" data-nxf-resource=\"nexus:\/\/gcri\/G-P08\/topics\/model-evaluation\/#map\" data-nxf-title=\"Artificial Intelligence and Autonomous Systems live map\" data-nxf-height=\"560\" hidden><\/div><\/div><\/div><\/div><\/div><\/section><section id=\"nx-s-4-sec\" class=\"nx-section\" data-hue=\"project\" aria-labelledby=\"nx-s-4\"><div class=\"nx-container\"><div class=\"nx-section-head\"><div class=\"nx-eyebrow-row\"><span class=\"nx-eyebrow\">03<\/span><span class=\"nx-rule\" aria-hidden=\"true\"><\/span><span class=\"nx-eyebrow\">Console<\/span><\/div><h2 id=\"nx-s-4\" class=\"nx-section-title\">Model registry, in outline.<\/h2><p class=\"nx-lede\">The working screen behind this subject: its structure, terms and states. Names are placeholders and values illustrative.<\/p><\/div><div class=\"nx-section-body\"><figure class=\"nx-fk nx-cx\" data-fk=\"cx-ai-registry\" data-cx=\"ai.registry\" data-variant=\"section\" aria-label=\"Model registry: illustrative console with placeholder names and values\"><div class=\"nx-fk-head\"><div class=\"nx-fk-heading\"><h3 class=\"nx-fk-title\">Model registry<\/h3><p class=\"nx-fk-sub\">Model evaluation \u00b7 Model registry \u00b7 Release gate<\/p><\/div><div class=\"nx-fk-actions\"><button type=\"button\" class=\"nx-fk-btn\" data-fk-act=\"menu\" aria-haspopup=\"true\" aria-expanded=\"false\" hidden><svg viewBox=\"0 0 24 24\" aria-hidden=\"true\" focusable=\"false\"><path d=\"M12 4v11\"\/><path d=\"m7 10 5 5 5-5\"\/><path d=\"M5 20h14\"\/><\/svg>Export<\/button><ul class=\"nx-fk-menu\" data-fk-menu=\"true\" hidden><li><button type=\"button\" data-fk-act=\"export-png\">Image<small>PNG of the map or chart<\/small><\/button><\/li><li><button type=\"button\" data-fk-act=\"export-csv\">Data<small>CSV of the figure's values<\/small><\/button><\/li><\/ul><\/div><\/div><div class=\"nx-cx-status\"><span class=\"nx-fk-pill\" data-state=\"near\"><i aria-hidden=\"true\"><\/i>One model blocked at gate<\/span><span class=\"nx-cx-prod\">GCRI \u00b7 Artificial Intelligence and Autonomous Systems<\/span><\/div><div class=\"nx-cx-grid\"><div class=\"nx-fk-panel nx-cx-p\" data-w=\"4\" data-t=\"table\"><div class=\"nx-fk-panel-head\"><div><h4 class=\"nx-fk-panel-title\">Models<\/h4><\/div><\/div><div class=\"nx-fk-tablewrap\" data-fk-view=\"table:cx-ai-registry-t0\"><table class=\"nx-fk-table\" data-fk-table=\"cx-ai-registry-t0\" aria-label=\"Models\"><thead><tr><th scope=\"col\" data-col=\"c0\" aria-sort=\"none\"><button type=\"button\" data-fk-sort=\"c0\">Model<\/button><\/th><th scope=\"col\" data-col=\"c1\" aria-sort=\"none\"><button type=\"button\" data-fk-sort=\"c1\">Use<\/button><\/th><th scope=\"col\" data-col=\"c2\" aria-sort=\"none\"><button type=\"button\" data-fk-sort=\"c2\">Risk tier<\/button><\/th><th scope=\"col\" data-col=\"c3\" aria-sort=\"none\"><button type=\"button\" data-fk-sort=\"c3\">Evaluation<\/button><\/th><th scope=\"col\" data-col=\"c4\" aria-sort=\"none\"><button type=\"button\" data-fk-sort=\"c4\">Gate<\/button><\/th><\/tr><\/thead><tbody><tr data-fk-key=\"cx-ai-registry:r0-0\" tabindex=\"0\" role=\"row\" aria-selected=\"false\" aria-current=\"true\" data-cur=\"true\"><td>M-184 triage assistant<\/td><td>Public service<\/td><td><span class=\"nx-fk-pill\" data-state=\"not\"><i aria-hidden=\"true\"><\/i>High<\/span><\/td><td>19 of 24 suites<\/td><td><span class=\"nx-fk-pill\" data-state=\"near\"><i aria-hidden=\"true\"><\/i>Blocked<\/span><\/td><\/tr><tr data-fk-key=\"cx-ai-registry:r0-1\" tabindex=\"0\" role=\"row\" aria-selected=\"false\"><td>M-320 flood nowcast<\/td><td>Science<\/td><td><span class=\"nx-fk-pill\" data-state=\"near\"><i aria-hidden=\"true\"><\/i>Medium<\/span><\/td><td>12 of 12<\/td><td><span class=\"nx-fk-pill\" data-state=\"met\"><i aria-hidden=\"true\"><\/i>Released<\/span><\/td><\/tr><tr data-fk-key=\"cx-ai-registry:r0-2\" tabindex=\"0\" role=\"row\" aria-selected=\"false\"><td>M-157 document search<\/td><td>Internal<\/td><td><span class=\"nx-fk-pill\" data-state=\"info\"><i aria-hidden=\"true\"><\/i>Low<\/span><\/td><td>8 of 8<\/td><td><span class=\"nx-fk-pill\" data-state=\"met\"><i aria-hidden=\"true\"><\/i>Released<\/span><\/td><\/tr><tr data-fk-key=\"cx-ai-registry:r0-3\" tabindex=\"0\" role=\"row\" aria-selected=\"false\"><td>M-563 agent planner<\/td><td>Agentic<\/td><td><span class=\"nx-fk-pill\" data-state=\"not\"><i aria-hidden=\"true\"><\/i>High<\/span><\/td><td>12 of 16<\/td><td><span class=\"nx-fk-pill\" data-state=\"info\"><i aria-hidden=\"true\"><\/i>Evaluating<\/span><\/td><\/tr><\/tbody><\/table><\/div><\/div><div class=\"nx-fk-panel nx-cx-p\" data-w=\"2\" data-t=\"kv\"><div class=\"nx-fk-panel-head\"><div><h4 class=\"nx-fk-panel-title\">Model card<\/h4><\/div><\/div><dl class=\"nx-fk-dl\"><div><dt>Intended use<\/dt><dd>triage support, human decides<\/dd><\/div><div><dt>Out of scope<\/dt><dd>eligibility decisions<\/dd><\/div><div><dt>Known limits<\/dt><dd>lower accuracy on rare cases<\/dd><\/div><div><dt>Owner<\/dt><dd>Model owner<\/dd><\/div><\/dl><\/div><div class=\"nx-fk-panel nx-cx-p\" data-w=\"3\" data-t=\"tl\"><div class=\"nx-fk-panel-head\"><div><h4 class=\"nx-fk-panel-title\">Incident log<\/h4><\/div><\/div><ol class=\"nx-cx-tl\"><li data-s=\"watch\"><span class=\"nx-cx-tt\">D-<span class=\"nx-cx-n\">21<\/span><\/span><span class=\"nx-cx-tag\">INC<\/span><span class=\"nx-cx-tx\">Hallucinated citation in summary<\/span><\/li><li data-s=\"ok\"><span class=\"nx-cx-tt\">D-<span class=\"nx-cx-n\">9<\/span><\/span><span class=\"nx-cx-tag\">MIT<\/span><span class=\"nx-cx-tx\">Retrieval grounding added<\/span><\/li><li data-s=\"alert\"><span class=\"nx-cx-tt\">D-<span class=\"nx-cx-n\">2<\/span><\/span><span class=\"nx-cx-tag\">INC<\/span><span class=\"nx-cx-tx\">Prompt injection via uploaded file<\/span><\/li><li data-s=\"live\"><span class=\"nx-cx-tt\">D-1<\/span><span class=\"nx-cx-tag\">MIT<\/span><span class=\"nx-cx-tx\">Tool permission narrowed<\/span><\/li><\/ol><\/div><div class=\"nx-fk-panel nx-cx-p\" data-w=\"3\" data-t=\"bullets\"><div class=\"nx-fk-panel-head\"><div><h4 class=\"nx-fk-panel-title\">Evaluation scores<\/h4><p class=\"nx-fk-panel-sub\">vs release threshold<\/p><\/div><\/div><div class=\"nx-fk-chart\"><svg class=\"nx-fk-svg\" viewBox=\"0 0 460 128\" width=\"460\" height=\"128\" role=\"img\" focusable=\"false\" preserveAspectRatio=\"xMidYMid meet\" aria-label=\"Evaluation scores\" data-fk-chart=\"bars\"><g data-m=\"row\" data-k=\"r-b0\"><text data-m=\"rlabel\" x=\"124\" y=\"19\" text-anchor=\"end\">Task accuracy<\/text><rect data-m=\"bar\" data-k=\"b-b0\" data-c=\"warn\" x=\"132\" y=\"4\" width=\"181.6\" height=\"22\"\/><text data-m=\"vlabel\" data-k=\"v-b0\" x=\"319.6\" y=\"19\">87 \u00b7 req 85<\/text><rect data-m=\"hit\" x=\"0\" y=\"0\" width=\"460\" height=\"30\"\/><\/g><g data-m=\"row\" data-k=\"r-b1\"><text data-m=\"rlabel\" x=\"124\" y=\"49\" text-anchor=\"end\">Robustness<\/text><rect data-m=\"bar\" data-k=\"b-b1\" data-c=\"warn\" x=\"132\" y=\"34\" width=\"165.9\" height=\"22\"\/><text data-m=\"vlabel\" data-k=\"v-b1\" x=\"303.9\" y=\"49\">80 \u00b7 req 75<\/text><rect data-m=\"hit\" x=\"0\" y=\"30\" width=\"460\" height=\"30\"\/><\/g><g data-m=\"row\" data-k=\"r-b2\"><text data-m=\"rlabel\" x=\"124\" y=\"79\" text-anchor=\"end\">Fairness gap (lower is better)<\/text><rect data-m=\"bar\" data-k=\"b-b2\" data-c=\"warn\" x=\"132\" y=\"64\" width=\"12.3\" height=\"22\"\/><text data-m=\"vlabel\" data-k=\"v-b2\" x=\"150.3\" y=\"79\">6 \u00b7 req 5<\/text><rect data-m=\"hit\" x=\"0\" y=\"60\" width=\"460\" height=\"30\"\/><\/g><g data-m=\"row\" data-k=\"r-b3\"><text data-m=\"rlabel\" x=\"124\" y=\"109\" text-anchor=\"end\">Refusal on misuse<\/text><rect data-m=\"bar\" data-k=\"b-b3\" data-c=\"neg\" x=\"132\" y=\"94\" width=\"176.5\" height=\"22\"\/><text data-m=\"vlabel\" data-k=\"v-b3\" x=\"314.5\" y=\"109\">85 \u00b7 req 95<\/text><rect data-m=\"hit\" x=\"0\" y=\"90\" width=\"460\" height=\"30\"\/><\/g><\/svg><\/div><\/div><\/div><div class=\"nx-fk-tip\" aria-hidden=\"true\" hidden><\/div><figcaption class=\"nx-fk-cap\"><span>Model registry: the screen's structure, terms and states for Model evaluation.<\/span><span data-honesty=\"true\">Placeholder names and illustrative values; a design, not a record or a live system.<\/span><\/figcaption><\/figure><\/div><\/div><\/section><section id=\"nx-sec-question\" class=\"nx-section\" data-hue=\"infrastructure\" data-tone=\"alt\" aria-labelledby=\"nx-s-5\"><div class=\"nx-container\"><div class=\"nx-section-head\"><div class=\"nx-eyebrow-row\"><span class=\"nx-eyebrow\">04<\/span><span class=\"nx-rule\" aria-hidden=\"true\"><\/span><span class=\"nx-eyebrow\">The question<\/span><\/div><h2 id=\"nx-s-5\" class=\"nx-section-title\">Model evaluation: how the risk develops.<\/h2><p class=\"nx-lede\">Validity, calibration and generalization. The dossier below sets out how the risk develops, what is measured, where the data come from, how it is modelled and who decides.<\/p><\/div><div class=\"nx-section-body\"><div class=\"nx-cells\" data-cols=\"2\"><div><div class=\"nx-glyph-row\"><span class=\"nx-glyph\" data-tone=\"accent\"><svg class=\"nx-i\" viewBox=\"0 0 24 24\" aria-hidden=\"true\" focusable=\"false\" fill=\"none\" stroke=\"currentColor\" stroke-width=\"1.5\" stroke-linecap=\"round\" stroke-linejoin=\"round\"><circle cx=\"12\" cy=\"12\" r=\"10\"\/><circle cx=\"12\" cy=\"12\" r=\"6\"\/><circle cx=\"12\" cy=\"12\" r=\"2\"\/><\/svg><\/span><span class=\"nx-eyebrow\">Mechanism<\/span><\/div><h3 class=\"nx-h4 nx-mt-2\">Benchmarks saturate and leak<\/h3><p class=\"nx-p nx-mt-1\">Models quickly reach high scores on fixed benchmarks, and test items can leak into training data. High scores then say more about the benchmark than about general capability.<\/p><\/div><div><div class=\"nx-glyph-row\"><span class=\"nx-glyph\" data-tone=\"accent\"><svg class=\"nx-i\" viewBox=\"0 0 24 24\" aria-hidden=\"true\" focusable=\"false\" fill=\"none\" stroke=\"currentColor\" stroke-width=\"1.5\" stroke-linecap=\"round\" stroke-linejoin=\"round\"><path d=\"m12.83 2.18a2 2 0 0 0-1.66 0L2.6 6.08a1 1 0 0 0 0 1.83l8.58 3.91a2 2 0 0 0 1.66 0l8.58-3.9a1 1 0 0 0 0-1.83Z\"\/><path d=\"m22 17.65-9.17 4.16a2 2 0 0 1-1.66 0L2 17.65\"\/><path d=\"m22 12.65-9.17 4.16a2 2 0 0 1-1.66 0L2 12.65\"\/><\/svg><\/span><span class=\"nx-eyebrow\">Mechanism<\/span><\/div><h3 class=\"nx-h4 nx-mt-2\">Test conditions differ from use<\/h3><p class=\"nx-p nx-mt-1\">Evaluations use set prompts, settings and tasks. Real users phrase things differently, combine tools and face new situations, so measured performance may not transfer.<\/p><\/div><div><div class=\"nx-glyph-row\"><span class=\"nx-glyph\" data-tone=\"accent\"><svg class=\"nx-i\" viewBox=\"0 0 24 24\" aria-hidden=\"true\" focusable=\"false\" fill=\"none\" stroke=\"currentColor\" stroke-width=\"1.5\" stroke-linecap=\"round\" stroke-linejoin=\"round\"><path d=\"m8 11 2 2 4-4\"\/><circle cx=\"11\" cy=\"11\" r=\"8\"\/><path d=\"m21 21-4.3-4.3\"\/><\/svg><\/span><span class=\"nx-eyebrow\">Mechanism<\/span><\/div><h3 class=\"nx-h4 nx-mt-2\">Adversarial testing finds hidden failures<\/h3><p class=\"nx-p nx-mt-1\">Red teaming and adversarial prompts reveal unsafe or unreliable behaviour that average-case metrics miss. Coverage depends on the testers' skill and the time allowed.<\/p><\/div><div><div class=\"nx-glyph-row\"><span class=\"nx-glyph\" data-tone=\"accent\"><svg class=\"nx-i\" viewBox=\"0 0 24 24\" aria-hidden=\"true\" focusable=\"false\" fill=\"none\" stroke=\"currentColor\" stroke-width=\"1.5\" stroke-linecap=\"round\" stroke-linejoin=\"round\"><path d=\"M22 12h-2.48a2 2 0 0 0-1.93 1.46l-2.35 8.36a.25.25 0 0 1-.48 0L9.24 2.18a.25.25 0 0 0-.48 0l-2.35 8.36A2 2 0 0 1 4.49 12H2\"\/><\/svg><\/span><span class=\"nx-eyebrow\">Mechanism<\/span><\/div><h3 class=\"nx-h4 nx-mt-2\">Performance drifts after deployment<\/h3><p class=\"nx-p nx-mt-1\">Model updates, changing inputs and new uses alter behaviour over time. Without monitoring, a model evaluated at release can behave differently months later.<\/p><\/div><\/div><\/div><\/div><\/section><section id=\"nx-sec-terms\" class=\"nx-section\" data-hue=\"project\" data-tone=\"alt\" data-variant=\"inset\" aria-labelledby=\"nx-s-6\"><div class=\"nx-container\"><div class=\"nx-section-head\"><div class=\"nx-eyebrow-row\"><span class=\"nx-eyebrow\">05<\/span><span class=\"nx-rule\" aria-hidden=\"true\"><\/span><span class=\"nx-eyebrow\">Key terms<\/span><\/div><h2 id=\"nx-s-6\" class=\"nx-section-title\">Terms used on this page.<\/h2><p class=\"nx-lede\">Each term as this page uses it. Where a standard-setting body defines the term, the wording follows that body.<\/p><\/div><div class=\"nx-section-body\"><div class=\"nx-table-wrap\"><table class=\"nx-table\" data-stack=\"true\"><thead><tr><th scope=\"col\">Term<\/th><th scope=\"col\">Definition<\/th><\/tr><\/thead><tbody><tr><td data-label=\"Term\">Benchmark<\/td><td data-label=\"Definition\">A fixed set of tasks and scoring rules used to compare model performance.<\/td><\/tr><tr><td data-label=\"Term\">Benchmark saturation<\/td><td data-label=\"Definition\">The point at which models reach scores so high that a benchmark no longer distinguishes between them.<\/td><\/tr><tr><td data-label=\"Term\">Calibration<\/td><td data-label=\"Definition\">The match between a model's stated confidence and how often it is actually correct.<\/td><\/tr><\/tbody><\/table><\/div><\/div><\/div><\/section><section id=\"nx-sec-indicators\" class=\"nx-section\" data-hue=\"infrastructure\" data-tone=\"alt\" data-variant=\"split\" aria-labelledby=\"nx-s-7\"><div class=\"nx-container\"><div class=\"nx-section-head\"><div class=\"nx-eyebrow-row\"><span class=\"nx-eyebrow\">06<\/span><span class=\"nx-rule\" aria-hidden=\"true\"><\/span><span class=\"nx-eyebrow\">Indicators<\/span><\/div><h2 id=\"nx-s-7\" class=\"nx-section-title\">What is measured, and how well.<\/h2><p class=\"nx-lede\">Each indicator carries its unit, its cadence, the kind of source behind it and the evidence label a value would carry. No values are shown: a programme supplies them under its own authority.<\/p><\/div><div class=\"nx-section-body\"><div class=\"nx-table-wrap\"><table class=\"nx-table\" data-stack=\"true\"><thead><tr><th scope=\"col\">Indicator<\/th><th scope=\"col\">Unit<\/th><th scope=\"col\">Cadence<\/th><th scope=\"col\">Source type<\/th><th scope=\"col\">Label<\/th><\/tr><\/thead><tbody><tr><td data-label=\"Indicator\">Benchmark score with uncertainty<\/td><td data-label=\"Unit\">score with confidence interval<\/td><td data-label=\"Cadence\">per evaluation<\/td><td data-label=\"Source type\">Published benchmark results<\/td><td data-label=\"Label\"><span class=\"nx-chip\">REPORTED<\/span><\/td><\/tr><tr><td data-label=\"Indicator\">Contamination checks reported<\/td><td data-label=\"Unit\">percent of evaluations reporting checks<\/td><td data-label=\"Cadence\">per evaluation<\/td><td data-label=\"Source type\">Evaluation reports<\/td><td data-label=\"Label\"><span class=\"nx-chip\">DECLARED<\/span><\/td><\/tr><tr><td data-label=\"Indicator\">Red-team findings by severity<\/td><td data-label=\"Unit\">count by severity class<\/td><td data-label=\"Cadence\">per exercise<\/td><td data-label=\"Source type\">Red-team reports<\/td><td data-label=\"Label\"><span class=\"nx-chip\">REPORTED<\/span><\/td><\/tr><tr><td data-label=\"Indicator\">Performance gap between subgroups<\/td><td data-label=\"Unit\">difference in error rates<\/td><td data-label=\"Cadence\">per evaluation<\/td><td data-label=\"Source type\">Disaggregated evaluation<\/td><td data-label=\"Label\"><span class=\"nx-chip\">OBSERVED<\/span><\/td><\/tr><tr><td data-label=\"Indicator\">Post-deployment incident reports<\/td><td data-label=\"Unit\">count by type<\/td><td data-label=\"Cadence\">monthly<\/td><td data-label=\"Source type\">Incident monitoring<\/td><td data-label=\"Label\"><span class=\"nx-chip\">REPORTED<\/span><\/td><\/tr><tr><td data-label=\"Indicator\">Calibration error<\/td><td data-label=\"Unit\">expected calibration error<\/td><td data-label=\"Cadence\">per evaluation<\/td><td data-label=\"Source type\">Evaluation on held-out data<\/td><td data-label=\"Label\"><span class=\"nx-chip\">OBSERVED<\/span><\/td><\/tr><\/tbody><\/table><\/div><div class=\"nx-live-slot\" data-nxf-slot=\"indicators\" data-nxf-kind=\"indicator\" data-nxf-resource=\"nexus:\/\/gcri\/G-P08\/topics\/model-evaluation\/#indicators\" data-nxf-title=\"Indicators\" data-nxf-height=\"520\" hidden><\/div><\/div><\/div><\/section><section id=\"nx-sec-pipeline\" class=\"nx-section\" data-hue=\"project\" data-tone=\"alt\" aria-labelledby=\"nx-s-8\"><div class=\"nx-container\"><div class=\"nx-section-head\"><div class=\"nx-eyebrow-row\"><span class=\"nx-eyebrow\">07<\/span><span class=\"nx-rule\" aria-hidden=\"true\"><\/span><span class=\"nx-eyebrow\">Canvas<\/span><\/div><h2 id=\"nx-s-8\" class=\"nx-section-title\">From sources to decisions.<\/h2><p class=\"nx-lede\">The topic as a computational graph: named sources feed the methods, an evidence review gates the outputs, and the decisions stay with the authorities that hold them.<\/p><\/div><div class=\"nx-section-body\"><div class=\"nx-plate\" data-wide=\"true\"><div class=\"nx-plate-label\"><span class=\"nx-eyebrow\"><strong>Pipeline canvas<\/strong> \u00b7 Model evaluation as a run graph<\/span><span class=\"nx-eyebrow\">Design, not a record<\/span><\/div><div class=\"nx-figure-frame\" data-scene=\"true\"><svg class=\"nx-figure nx-ui nx-dag\" viewBox=\"0 0 1200 564\" data-nx-dag=\"true\" data-w=\"1200\" data-h=\"564\" role=\"img\" aria-labelledby=\"nx-dag-g-p08-topics-model-evaluation-ai-t nx-dag-g-p08-topics-model-evaluation-ai-d\" data-nx-z=\"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\"><title id=\"nx-dag-g-p08-topics-model-evaluation-ai-t\">Model evaluation: AI risk management, govern to manage<\/title><desc id=\"nx-dag-g-p08-topics-model-evaluation-ai-d\">Governance sets roles and policies; the context and intended use are mapped; capability, safety, robustness and fairness are measured; risks are managed with mitigations and a deployment gate; incidents feed monitoring and the registry. Design of the process, not a record of a run.<\/desc><\/svg><\/div><\/div><p class=\"nx-caption\"><strong>Fig. 02<\/strong><span>Nodes carry their type, two metric rows and a state; connectors show what feeds what; dashed frames name the owners; the timeline is the cadence. Design of the pipeline, not a record of a run.<\/span><\/p><\/div><\/div><\/section><section id=\"nx-sec-sources\" class=\"nx-section\" data-hue=\"infrastructure\" data-tone=\"alt\" data-variant=\"stack\" aria-labelledby=\"nx-s-9\"><div class=\"nx-container\"><div class=\"nx-section-head\"><div class=\"nx-eyebrow-row\"><span class=\"nx-eyebrow\">08<\/span><span class=\"nx-rule\" aria-hidden=\"true\"><\/span><span class=\"nx-eyebrow\">Sources<\/span><\/div><h2 id=\"nx-s-9\" class=\"nx-section-title\">Where the evidence comes from.<\/h2><p class=\"nx-lede\">Named public datasets and institutional records, each with its licence class. Sources by agreement are used only under their terms and never republished.<\/p><\/div><div class=\"nx-section-body\"><div class=\"nx-table-wrap\"><table class=\"nx-table\" data-stack=\"true\"><thead><tr><th scope=\"col\">Source<\/th><th scope=\"col\">What it gives<\/th><th scope=\"col\">Licence class<\/th><\/tr><\/thead><tbody><tr><td data-label=\"Source\">Stanford HELM (Holistic Evaluation of Language Models)<\/td><td data-label=\"What it gives\">Open, multi-metric evaluations of language models<\/td><td data-label=\"Licence class\">Open source, Apache 2.0 for code; results public<\/td><\/tr><tr><td data-label=\"Source\">MLCommons benchmarks, including AILuminate<\/td><td data-label=\"What it gives\">Standardised performance and safety benchmarks<\/td><td data-label=\"Licence class\">Public results, MLCommons terms (to check)<\/td><\/tr><tr><td data-label=\"Source\">International AI Safety Report<\/td><td data-label=\"What it gives\">Synthesis of evidence on capabilities, risks and evaluation limits<\/td><td data-label=\"Licence class\">Public, terms per publisher<\/td><\/tr><tr><td data-label=\"Source\">Model cards and system cards from developers<\/td><td data-label=\"What it gives\">Developer-reported evaluations and limitations<\/td><td data-label=\"Licence class\">Public documents, terms per developer<\/td><\/tr><tr><td data-label=\"Source\">Epoch AI benchmark and model databases<\/td><td data-label=\"What it gives\">Compiled benchmark results and model characteristics<\/td><td data-label=\"Licence class\">Open, CC BY 4.0<\/td><\/tr><\/tbody><\/table><\/div><div class=\"nx-live-slot\" data-nxf-slot=\"evidence\" data-nxf-kind=\"evidence\" data-nxf-resource=\"nexus:\/\/gcri\/G-P08\/topics\/model-evaluation\/#evidence\" data-nxf-title=\"Sources\" data-nxf-height=\"520\" hidden><\/div><\/div><\/div><\/section><section id=\"nx-sec-decisions\" class=\"nx-section\" data-hue=\"project\" data-tone=\"alt\" aria-labelledby=\"nx-s-10\"><div class=\"nx-container\"><div class=\"nx-section-head\"><div class=\"nx-eyebrow-row\"><span class=\"nx-eyebrow\">09<\/span><span class=\"nx-rule\" aria-hidden=\"true\"><\/span><span class=\"nx-eyebrow\">Decisions<\/span><\/div><h2 id=\"nx-s-10\" class=\"nx-section-title\">Who decides, and what the evidence serves.<\/h2><p class=\"nx-lede\">Artificial Intelligence and Autonomous Systems prepares evidence. The decisions below are taken by the authorities named, under their own rules, and never by the platform.<\/p><\/div><div class=\"nx-section-body\" data-lv=\"ledger\"><div class=\"nx-table-wrap\"><table class=\"nx-table\" data-stack=\"true\"><thead><tr><th scope=\"col\">Who decides<\/th><th scope=\"col\">The question<\/th><th scope=\"col\">Evidence this page serves<\/th><\/tr><\/thead><tbody><tr><td data-label=\"Who decides\">Developers<\/td><td data-label=\"The question\">Whether a model is ready for release and what limits to state<\/td><td data-label=\"Evidence this page serves\">Evaluation protocols; red-team findings<\/td><\/tr><tr><td data-label=\"Who decides\">Deploying organizations<\/td><td data-label=\"The question\">Whether a model is fit for a specific use<\/td><td data-label=\"Evidence this page serves\">Use-specific evaluations; subgroup performance<\/td><\/tr><tr><td data-label=\"Who decides\">Regulators and AI safety institutes<\/td><td data-label=\"The question\">What evaluation evidence to require<\/td><td data-label=\"Evidence this page serves\">Evaluation standards and reporting templates<\/td><\/tr><\/tbody><\/table><\/div><h3 class=\"nx-h4 nx-mt-4\">Outputs this topic produces<\/h3><ul class=\"nx-list\"><li>Evaluation protocols with uncertainty and contamination checks<\/li><li>Disaggregated evaluation designs for subgroup performance<\/li><li>Red-team methods for public sector use cases<\/li><li>Evaluation reporting templates<\/li><\/ul><\/div><\/div><\/section><section id=\"nx-sec-core-questions\" class=\"nx-section\" data-hue=\"infrastructure\" data-tone=\"alt\" data-variant=\"inset\" aria-labelledby=\"nx-s-11\"><div class=\"nx-container\"><div class=\"nx-section-head\"><div class=\"nx-eyebrow-row\"><span class=\"nx-eyebrow\">10<\/span><span class=\"nx-rule\" aria-hidden=\"true\"><\/span><span class=\"nx-eyebrow\">Core questions<\/span><\/div><h2 id=\"nx-s-11\" class=\"nx-section-title\">What evaluators ask.<\/h2><\/div><div class=\"nx-section-body\" data-lv=\"columns\"><ul class=\"nx-list\"><li>Does the test measure the capability or property that matters for the intended use (construct validity)?<\/li><li>When the model reports high confidence, is it right as often as that confidence implies (calibration)?<\/li><li>Does performance hold on data from different places, times or populations (generalization)?<\/li><li>Has test data leaked into training, inflating results (contamination)?<\/li><\/ul><\/div><\/div><\/section><section id=\"nx-sec-mechanisms\" class=\"nx-section\" data-hue=\"project\" data-variant=\"split\" aria-labelledby=\"nx-s-12\"><div class=\"nx-container\"><div class=\"nx-section-head\"><div class=\"nx-eyebrow-row\"><span class=\"nx-eyebrow\">11<\/span><span class=\"nx-rule\" aria-hidden=\"true\"><\/span><span class=\"nx-eyebrow\">Mechanisms<\/span><\/div><h2 id=\"nx-s-12\" class=\"nx-section-title\">Why scores mislead.<\/h2><\/div><div class=\"nx-section-body\"><p class=\"nx-p\">Models learn statistical regularities in their training data, including spurious ones. A medical imaging model may learn the hospital's scanner signature rather than the disease; a language model may recognise benchmark questions it saw during training. When deployment data differ from test data, a problem known as distribution shift, performance can fall sharply and without warning.<\/p><p class=\"nx-p nx-mt-2\">Large general-purpose models add new difficulties. Their capabilities are broad and emergent, outputs vary with phrasing, and the line between capability and memorisation is blurred. Aggregate scores can hide failures concentrated in particular groups or tasks.<\/p><\/div><\/div><\/section><section id=\"nx-sec-methods-and-data\" class=\"nx-section\" data-hue=\"infrastructure\" data-tone=\"alt\" data-variant=\"stack\" aria-labelledby=\"nx-s-13\"><div class=\"nx-container\"><div class=\"nx-section-head\"><div class=\"nx-eyebrow-row\"><span class=\"nx-eyebrow\">12<\/span><span class=\"nx-rule\" aria-hidden=\"true\"><\/span><span class=\"nx-eyebrow\">Methods and data<\/span><\/div><h2 id=\"nx-s-13\" class=\"nx-section-title\">How evaluation is done, and its limits.<\/h2><\/div><div class=\"nx-section-body\"><ol class=\"nx-rows\"><li><span class=\"nx-eyebrow nx-index\">01<\/span><div><h3 class=\"nx-h4\">Held-out and out-of-distribution testing<\/h3><p class=\"nx-p nx-mt-1\">Testing on data deliberately different from training, analysed with <a class=\"nx-link\" href=\"https:\/\/therisk.global\/ai-autonomous-systems\/stem\/probability-statistics\/\">probability and statistics<\/a>. No test set covers every deployment condition.<\/p><\/div><\/li><li><span class=\"nx-eyebrow nx-index\">02<\/span><div><h3 class=\"nx-h4\">Calibration analysis<\/h3><p class=\"nx-p nx-mt-1\">Comparing stated confidence with observed accuracy, and checking how it changes under shift.<\/p><\/div><\/li><li><span class=\"nx-eyebrow nx-index\">03<\/span><div><h3 class=\"nx-h4\">Disaggregated evaluation<\/h3><p class=\"nx-p nx-mt-1\">Reporting performance by subgroup and task rather than only on average. Small subgroups give uncertain estimates.<\/p><\/div><\/li><li><span class=\"nx-eyebrow nx-index\">04<\/span><div><h3 class=\"nx-h4\">Human and expert evaluation<\/h3><p class=\"nx-p nx-mt-1\">Structured assessment of open-ended outputs, using <a class=\"nx-link\" href=\"https:\/\/therisk.global\/ai-autonomous-systems\/stem\/social-science\/\">social and behavioural science<\/a> methods for rater agreement. Human judgement is costly and can be inconsistent.<\/p><\/div><\/li><\/ol><\/div><\/div><\/section><section id=\"nx-sec-methods\" class=\"nx-section\" data-hue=\"project\" aria-labelledby=\"nx-s-14\"><div class=\"nx-container\"><div class=\"nx-section-head\"><div class=\"nx-eyebrow-row\"><span class=\"nx-eyebrow\">13<\/span><span class=\"nx-rule\" aria-hidden=\"true\"><\/span><span class=\"nx-eyebrow\">Methods<\/span><\/div><h2 id=\"nx-s-14\" class=\"nx-section-title\">How it is studied, and what the methods cannot say.<\/h2><\/div><div class=\"nx-section-body\"><div class=\"nx-methods\" data-limits=\"true\"><ol class=\"nx-mledger\" aria-label=\"Methods\"><li><a class=\"nx-mledger-main\" href=\"https:\/\/therisk.global\/ai-autonomous-systems\/stem\/ai-assurance\/\"><span class=\"nx-mledger-n\" aria-hidden=\"true\">01<\/span><span class=\"nx-mledger-body\"><span class=\"nx-mledger-name\">AI evaluation and assurance engineering<\/span><span class=\"nx-mledger-role\">Assurance cases for AI systems \u00b7 Structured red teaming \u00b7 Post-deployment monitoring<\/span><\/span><span class=\"nx-mledger-go\"><span class=\"nx-mledger-go-label\">Applied here<\/span><svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\" fill=\"none\" stroke=\"currentColor\" stroke-width=\"1.5\" stroke-linecap=\"round\" stroke-linejoin=\"round\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/span><\/a><div class=\"nx-mledger-aside\"><a class=\"nx-mledger-canon\" href=\"https:\/\/therisk.global\/stem\/engineering\/ai-assurance\/\" aria-label=\"AI evaluation and assurance engineering: canonical page on the STEM Commons\">Canonical \u00b7 STEM Commons <span aria-hidden=\"true\">\u2197<\/span><\/a><\/div><\/li><li><a class=\"nx-mledger-main\" href=\"https:\/\/therisk.global\/ai-autonomous-systems\/stem\/probability-statistics\/\"><span class=\"nx-mledger-n\" aria-hidden=\"true\">02<\/span><span class=\"nx-mledger-body\"><span class=\"nx-mledger-name\">Probability and statistics<\/span><span class=\"nx-mledger-role\">Uncertainty on benchmark scores \u00b7 Calibration assessment \u00b7 Audit sampling<\/span><\/span><span class=\"nx-mledger-go\"><span class=\"nx-mledger-go-label\">Applied here<\/span><svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\" fill=\"none\" stroke=\"currentColor\" stroke-width=\"1.5\" stroke-linecap=\"round\" stroke-linejoin=\"round\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/span><\/a><div class=\"nx-mledger-aside\"><a class=\"nx-mledger-canon\" href=\"https:\/\/therisk.global\/stem\/mathematics\/probability-statistics\/\" aria-label=\"Probability and statistics: canonical page on the STEM Commons\">Canonical \u00b7 STEM Commons <span aria-hidden=\"true\">\u2197<\/span><\/a><\/div><\/li><li><a class=\"nx-mledger-main\" href=\"https:\/\/therisk.global\/ai-autonomous-systems\/stem\/ai-ml\/\"><span class=\"nx-mledger-n\" aria-hidden=\"true\">03<\/span><span class=\"nx-mledger-body\"><span class=\"nx-mledger-name\">Artificial intelligence and machine learning<\/span><span class=\"nx-mledger-role\">Distribution shift analysis \u00b7 Fine-tuning for a specific use \u00b7 Interpretability methods<\/span><\/span><span class=\"nx-mledger-go\"><span class=\"nx-mledger-go-label\">Applied here<\/span><svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\" fill=\"none\" stroke=\"currentColor\" stroke-width=\"1.5\" stroke-linecap=\"round\" stroke-linejoin=\"round\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/span><\/a><div class=\"nx-mledger-aside\"><a class=\"nx-mledger-canon\" href=\"https:\/\/therisk.global\/stem\/technology\/ai-ml\/\" aria-label=\"Artificial intelligence and machine learning: canonical page on the STEM Commons\">Canonical \u00b7 STEM Commons <span aria-hidden=\"true\">\u2197<\/span><\/a><\/div><\/li><li><a class=\"nx-mledger-main\" href=\"https:\/\/therisk.global\/ai-autonomous-systems\/stem\/social-science\/\"><span class=\"nx-mledger-n\" aria-hidden=\"true\">04<\/span><span class=\"nx-mledger-body\"><span class=\"nx-mledger-name\">Social and behavioural science<\/span><span class=\"nx-mledger-role\">Human oversight studies \u00b7 Affected community research \u00b7 Workplace adoption studies<\/span><\/span><span class=\"nx-mledger-go\"><span class=\"nx-mledger-go-label\">Applied here<\/span><svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\" fill=\"none\" stroke=\"currentColor\" stroke-width=\"1.5\" stroke-linecap=\"round\" stroke-linejoin=\"round\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/span><\/a><div class=\"nx-mledger-aside\"><a class=\"nx-mledger-canon\" href=\"https:\/\/therisk.global\/stem\/science\/social-science\/\" aria-label=\"Social and behavioural science: canonical page on the STEM Commons\">Canonical \u00b7 STEM Commons <span aria-hidden=\"true\">\u2197<\/span><\/a><\/div><\/li><\/ol><aside class=\"nx-limits\" aria-label=\"What the methods cannot say\"><span class=\"nx-eyebrow\">What the methods cannot say<\/span><p class=\"nx-limits-text\">No evaluation predicts all behaviour in deployment. Benchmark results depend on prompts, settings and scoring choices that are not always reported. Red teaming finds some failures, never all. Developer-reported results are not independent. Disaggregated evaluation needs demographic data that may be unavailable or sensitive. This platform rates and approves no model; deployers and regulators decide.<\/p><\/aside><\/div><\/div><\/div><\/section><section id=\"nx-s-15-sec\" class=\"nx-section\" data-hue=\"infrastructure\" data-tone=\"alt\" aria-labelledby=\"nx-s-15\"><div class=\"nx-container\"><div class=\"nx-section-head\"><div class=\"nx-eyebrow-row\"><span class=\"nx-eyebrow\">14<\/span><span class=\"nx-rule\" aria-hidden=\"true\"><\/span><span class=\"nx-eyebrow\">Connected<\/span><\/div><h2 id=\"nx-s-15\" class=\"nx-section-title\">Where this page leads across the Consortium.<\/h2><\/div><div class=\"nx-section-body\"><ul class=\"nx-dest\"><li><a class=\"nx-dest-card\" href=\"https:\/\/globalriskforum.com\/ai-digital-governance\/topics\/ai-oversight-and-accountability\/\" target=\"_blank\" rel=\"noopener\"><span class=\"nx-dest-top\"><span class=\"nx-dest-rel\">Related topic<\/span><span class=\"nx-dest-host\">GRF<\/span><\/span><span class=\"nx-dest-title\">AI oversight and accountability<\/span><span class=\"nx-dest-why\">Scientific model evaluation informs institutional governance and oversight debates.<\/span><span class=\"nx-dest-go\" aria-hidden=\"true\"><svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\" fill=\"none\" stroke=\"currentColor\" stroke-width=\"1.5\" stroke-linecap=\"round\" stroke-linejoin=\"round\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/span><\/a><\/li><li><a class=\"nx-dest-card\" href=\"https:\/\/globalriskalliance.com\/ai-financial-model-risk\/topics\/model-risk-governance\/\" target=\"_blank\" rel=\"noopener\"><span class=\"nx-dest-top\"><span class=\"nx-dest-rel\">Related topic<\/span><span class=\"nx-dest-host\">GRA<\/span><\/span><span class=\"nx-dest-title\">Model risk governance<\/span><span class=\"nx-dest-why\">AI testing methods connect to financial-sector model validation without substituting for it.<\/span><span class=\"nx-dest-go\" aria-hidden=\"true\"><svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\" fill=\"none\" stroke=\"currentColor\" stroke-width=\"1.5\" stroke-linecap=\"round\" stroke-linejoin=\"round\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/span><\/a><\/li><\/ul><\/div><\/div><\/section><section id=\"nx-s-16-sec\" class=\"nx-section\" data-hue=\"project\" aria-labelledby=\"nx-s-16\"><div class=\"nx-container\"><div class=\"nx-section-head\"><div class=\"nx-eyebrow-row\"><span class=\"nx-eyebrow\">15<\/span><span class=\"nx-rule\" aria-hidden=\"true\"><\/span><span class=\"nx-eyebrow\">Questions<\/span><\/div><h2 id=\"nx-s-16\" class=\"nx-section-title\">Questions and answers.<\/h2><\/div><div class=\"nx-section-body\"><div class=\"nx-faq\"><details><summary>Why are benchmark scores not enough?<\/summary><div class=\"nx-answer\">Because models can be trained on test items, benchmarks saturate, and test conditions differ from real use. Use-specific, adversarial and post-deployment evaluation are needed as well.<\/div><\/details><details><summary>What is red teaming in AI?<\/summary><div class=\"nx-answer\">Structured adversarial testing in which people or automated tools try to make a model produce harmful, insecure or unreliable outputs, to find failures before deployment.<\/div><\/details><details><summary>What is data contamination?<\/summary><div class=\"nx-answer\">The presence of benchmark test items in a model's training data, which inflates scores and makes results unreliable as measures of general capability.<\/div><\/details><details><summary>Is a high benchmark score evidence a model is safe to deploy?<\/summary><div class=\"nx-answer\">No. It shows performance on that test under its conditions. Fitness for a use needs evaluation on representative data, with failure analysis and monitoring after deployment.<\/div><\/details><details><summary>What is benchmark saturation?<\/summary><div class=\"nx-answer\">When models reach near-perfect scores, the benchmark no longer discriminates between them, and further gains may reflect overfitting to the test rather than real capability.<\/div><\/details><details><summary>Why evaluate after deployment too?<\/summary><div class=\"nx-answer\">Because data, users and the model itself can change. Monitoring detects drift and new failure modes that pre-release testing could not anticipate.<\/div><\/details><details><summary>Will the platform publish evaluation protocols?<\/summary><div class=\"nx-answer\">Protocols are designed to appear on <a class=\"nx-link\" href=\"https:\/\/therisk.global\/ai-autonomous-systems\/tools\/models\/\">models<\/a> and in publications once reviewed, each with its scope and limits.<\/div><\/details><\/div><\/div><\/div><\/section><section class=\"nx-section\" data-tone=\"alt\" aria-labelledby=\"nx-rel-17\"><div class=\"nx-container\"><span class=\"nx-eyebrow\">Related<\/span><h2 id=\"nx-rel-17\" class=\"nx-h3 nx-mt-2\">Around Artificial Intelligence and Autonomous Systems.<\/h2><nav class=\"nx-related nx-mt-3\" aria-label=\"Related\"><a href=\"https:\/\/therisk.global\/ai-autonomous-systems\/topics\/agentic-systems\/\">Agentic systems <svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\" fill=\"none\" stroke=\"currentColor\" stroke-width=\"1.5\" stroke-linecap=\"round\" stroke-linejoin=\"round\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><a href=\"https:\/\/therisk.global\/ai-autonomous-systems\/topics\/ai-cyber-security\/\">AI cyber security <svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\" fill=\"none\" stroke=\"currentColor\" stroke-width=\"1.5\" stroke-linecap=\"round\" stroke-linejoin=\"round\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><a href=\"https:\/\/therisk.global\/ai-autonomous-systems\/topics\/ai-in-science\/\">AI in science <svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\" fill=\"none\" stroke=\"currentColor\" stroke-width=\"1.5\" stroke-linecap=\"round\" stroke-linejoin=\"round\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><a href=\"https:\/\/therisk.global\/ai-autonomous-systems\/topics\/ai-and-public-services\/\">AI and public services <svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\" fill=\"none\" stroke=\"currentColor\" stroke-width=\"1.5\" stroke-linecap=\"round\" stroke-linejoin=\"round\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><a href=\"https:\/\/therisk.global\/ai-autonomous-systems\/topics\/autonomous-physical-systems\/\">Autonomous physical systems <svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\" fill=\"none\" stroke=\"currentColor\" stroke-width=\"1.5\" stroke-linecap=\"round\" stroke-linejoin=\"round\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><a href=\"https:\/\/therisk.global\/ai-autonomous-systems\/topics\/compute-and-energy\/\">Compute and energy <svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\" fill=\"none\" stroke=\"currentColor\" stroke-width=\"1.5\" stroke-linecap=\"round\" stroke-linejoin=\"round\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><a href=\"https:\/\/therisk.global\/ai-autonomous-systems\/\">Artificial Intelligence and Autonomous Systems home <svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\" fill=\"none\" stroke=\"currentColor\" stroke-width=\"1.5\" stroke-linecap=\"round\" stroke-linejoin=\"round\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><a href=\"https:\/\/therisk.global\/ai-autonomous-systems\/counterparts\/\">Counterparts across the Consortium <svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\" fill=\"none\" stroke=\"currentColor\" stroke-width=\"1.5\" stroke-linecap=\"round\" stroke-linejoin=\"round\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><a href=\"https:\/\/therisk.global\/\">The Global Centre for Risk and Innovation home <svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\" fill=\"none\" stroke=\"currentColor\" stroke-width=\"1.5\" stroke-linecap=\"round\" stroke-linejoin=\"round\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><a href=\"https:\/\/therisk.global\/trust\/\">GCRI governance and trust <svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\" fill=\"none\" stroke=\"currentColor\" stroke-width=\"1.5\" stroke-linecap=\"round\" stroke-linejoin=\"round\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><\/nav><\/div><\/section><section class=\"nx-closing\" aria-labelledby=\"nx-close-18\"><div class=\"nx-container\"><span class=\"nx-eyebrow\">Artificial Intelligence and Autonomous Systems<\/span><h2 id=\"nx-close-18\" class=\"nx-section-title\">Bring your expertise to this field.<\/h2><p class=\"nx-lede\">The current service level is R0: information and enquiries. An enquiry receives a reference and names the next step; applications, payments, compute allocations and registrations open only when their team, terms and review capacity exist.<\/p><div class=\"nx-actions\"><a class=\"nx-btn\" data-variant=\"primary\" href=\"mailto:contact@therisk.global?subject=Artificial%20Intelligence%20and%20Autonomous%20Systems%3A%20Model%20evaluation\">Write to the team <svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\" fill=\"none\" stroke=\"currentColor\" stroke-width=\"1.5\" stroke-linecap=\"round\" stroke-linejoin=\"round\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><a class=\"nx-btn\" data-variant=\"secondary\" href=\"https:\/\/therisk.global\/ai-autonomous-systems\/contact\/\">See all contact routes<\/a><\/div><\/div><\/section><\/div><\/div><script class=\"_iub_cs_skip\">window.__nxMount ? window.__nxMount(document.currentScript) : null;<\/script>\n\n\n<script type=\"application\/ld+json\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"FAQPage\",\"@id\":\"https:\/\/therisk.global\/ai-autonomous-systems\/topics\/model-evaluation\/#faq\",\"isPartOf\":{\"@id\":\"https:\/\/therisk.global\/ai-autonomous-systems\/topics\/model-evaluation\/#webpage\"},\"inLanguage\":\"en\",\"mainEntity\":[{\"@type\":\"Question\",\"name\":\"Why are benchmark scores not enough?\",\"acceptedAnswer\":{\"@type\":\"Answer\",\"text\":\"Because models can be trained on test items, benchmarks saturate, and test conditions differ from real use. Use-specific, adversarial and post-deployment evaluation are needed as well.\"}},{\"@type\":\"Question\",\"name\":\"What is red teaming in AI?\",\"acceptedAnswer\":{\"@type\":\"Answer\",\"text\":\"Structured adversarial testing in which people or automated tools try to make a model produce harmful, insecure or unreliable outputs, to find failures before deployment.\"}},{\"@type\":\"Question\",\"name\":\"What is data contamination?\",\"acceptedAnswer\":{\"@type\":\"Answer\",\"text\":\"The presence of benchmark test items in a model's training data, which inflates scores and makes results unreliable as measures of general capability.\"}},{\"@type\":\"Question\",\"name\":\"Is a high benchmark score evidence a model is safe to deploy?\",\"acceptedAnswer\":{\"@type\":\"Answer\",\"text\":\"No. It shows performance on that test under its conditions. Fitness for a use needs evaluation on representative data, with failure analysis and monitoring after deployment.\"}},{\"@type\":\"Question\",\"name\":\"What is benchmark saturation?\",\"acceptedAnswer\":{\"@type\":\"Answer\",\"text\":\"When models reach near-perfect scores, the benchmark no longer discriminates between them, and further gains may reflect overfitting to the test rather than real capability.\"}},{\"@type\":\"Question\",\"name\":\"Why evaluate after deployment too?\",\"acceptedAnswer\":{\"@type\":\"Answer\",\"text\":\"Because data, users and the model itself can change. Monitoring detects drift and new failure modes that pre-release testing could not anticipate.\"}},{\"@type\":\"Question\",\"name\":\"Will the platform publish evaluation protocols?\",\"acceptedAnswer\":{\"@type\":\"Answer\",\"text\":\"Protocols are designed to appear on models and in publications once reviewed, each with its scope and limits.\"}}]},{\"@type\":\"DefinedTermSet\",\"@id\":\"https:\/\/therisk.global\/ai-autonomous-systems\/topics\/model-evaluation\/#terms\",\"name\":\"Key terms: Model evaluation\",\"inLanguage\":\"en\",\"isPartOf\":{\"@id\":\"https:\/\/therisk.global\/ai-autonomous-systems\/topics\/model-evaluation\/#webpage\"},\"hasDefinedTerm\":[{\"@type\":\"DefinedTerm\",\"name\":\"Benchmark\",\"description\":\"A fixed set of tasks and scoring rules used to compare model performance.\",\"inDefinedTermSet\":{\"@id\":\"https:\/\/therisk.global\/ai-autonomous-systems\/topics\/model-evaluation\/#terms\"}},{\"@type\":\"DefinedTerm\",\"name\":\"Benchmark saturation\",\"description\":\"The point at which models reach scores so high that a benchmark no longer distinguishes between them.\",\"inDefinedTermSet\":{\"@id\":\"https:\/\/therisk.global\/ai-autonomous-systems\/topics\/model-evaluation\/#terms\"}},{\"@type\":\"DefinedTerm\",\"name\":\"Calibration\",\"description\":\"The match between a model's stated confidence and how often it is actually correct.\",\"inDefinedTermSet\":{\"@id\":\"https:\/\/therisk.global\/ai-autonomous-systems\/topics\/model-evaluation\/#terms\"}}]}]}<\/script>\n\n<!-- nx-rev:7127c937dada8767 -->","protected":false},"excerpt":{"rendered":"<p>LightDarkAuto G-P08 \u00b7 TOPIC \u00b7 model-evaluationGCRI dossier \u00b7 service level R0 GCRI\u00b7Artificial Intelligence and Autonomous Systems\u00b7In development Topic dossier \u00b7 Artificial Intelligence and Autonomous Systems Model evaluation Model evaluation asks &hellip; <a href=\"https:\/\/therisk.global\/ai-autonomous-systems\/topics\/model-evaluation\/\" class=\"more-link\">Continue reading<span class=\"screen-reader-text\"> &#8220;Model evaluation&#8221;<\/span><\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"parent":78,"menu_order":320,"comment_status":"closed","ping_status":"closed","template":"elementor_header_footer","meta":{"_buddyx_sub_header_visibility":"","_buddyx_sub_header_title_visibility":"","_hide_show_side_panel":"","_buddyxpro_page_sidebar":"","_buddyxpro_page_disable_header":"","_buddyxpro_page_disable_footer":"","_buddyxpro_page_content_width":"","_buddyxpro_page_header_style":"","_buddyxpro_page_color_mode":"","_buddyxpro_page_loader":"","footnotes":""},"class_list":["post-105","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/therisk.global\/ai-autonomous-systems\/wp-json\/wp\/v2\/pages\/105","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/therisk.global\/ai-autonomous-systems\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/therisk.global\/ai-autonomous-systems\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/therisk.global\/ai-autonomous-systems\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/therisk.global\/ai-autonomous-systems\/wp-json\/wp\/v2\/comments?post=105"}],"version-history":[{"count":2,"href":"https:\/\/therisk.global\/ai-autonomous-systems\/wp-json\/wp\/v2\/pages\/105\/revisions"}],"predecessor-version":[{"id":229,"href":"https:\/\/therisk.global\/ai-autonomous-systems\/wp-json\/wp\/v2\/pages\/105\/revisions\/229"}],"up":[{"embeddable":true,"href":"https:\/\/therisk.global\/ai-autonomous-systems\/wp-json\/wp\/v2\/pages\/78"}],"wp:attachment":[{"href":"https:\/\/therisk.global\/ai-autonomous-systems\/wp-json\/wp\/v2\/media?parent=105"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}