{"id":39,"date":"2026-11-01T22:11:58","date_gmt":"2026-11-01T22:11:58","guid":{"rendered":"https:\/\/therisk.global\/ai-autonomous-systems\/?page_id=39"},"modified":"2026-11-01T22:11:58","modified_gmt":"2026-11-01T22:11:58","slug":"ai-ml","status":"publish","type":"page","link":"https:\/\/therisk.global\/ai-autonomous-systems\/stem\/ai-ml\/","title":{"rendered":"Machine learning behaviour"},"content":{"rendered":"\n<!-- Nexus network page \u00b7 https:\/\/therisk.global\/ai-autonomous-systems\/stem\/ai-ml\/ \u00b7 source template \u00b7 generated by integrations\/wordpress\/network\/build.mjs, never edit this file --><script>window.nexusLiveStrings={\"label\":\"Load the live map\",\"note\":\"Imagery and feeds come from other services (NASA, USGS, OpenStreetMap) and load only when you ask.\"};<\/script><script>(function(){\"use strict\";var n=window,l=document,y=\"nexus.live-map\",$=\"nexus.live-map.reload\";if(n.NXLive){return}var E=[\/(^|\\.)gibs\\.earthdata\\.nasa\\.gov$\/i,\/(^|\\.)eonet\\.gsfc\\.nasa\\.gov$\/i,\/(^|\\.)power\\.larc\\.nasa\\.gov$\/i,\/(^|\\.)earthquake\\.usgs\\.gov$\/i,\/(^|\\.)open-meteo\\.com$\/i,\/(^|\\.)openfreemap\\.org$\/i,\/(^|\\.)openstreetmap\\.org$\/i,\/(^|\\.)cartocdn\\.com$\/i,\/(^|\\.)arcgisonline\\.com$\/i,\/(^|\\.)maplibre\\.org$\/i,\/(^|\\.)protomaps\\.com$\/i,\/(^|\\.)versatiles\\.org$\/i,\/(^|\\.)amazonaws\\.com$\/i,\/(^|\\.)raw\\.githubusercontent\\.com$\/i,\/(^|\\.)media\\.githubusercontent\\.com$\/i,\/(^|\\.)geoboundaries\\.org$\/i,\/(^|\\.)jsdelivr\\.net$\/i,\/(^|\\.)unpkg\\.com$\/i,\/(^|\\.)cdn[j]s\\.cloudflare\\.com$\/i,\/(^|\\.)googleapis\\.com$\/i,\/(^|\\.)gstatic\\.com$\/i];var C=[\/(^|\\.)amazonaws\\.com$\/i];var b=n.nexusLiveStrings?n.nexusLiveStrings:{};var A={label:b.label?b.label:\"Load the live map\",note:b.note?b.note:\"Imagery and feeds come from other services (NASA, USGS, OpenStreetMap) and load only when you ask.\"};var I=[\"experience\",\"external\",\"optional-content\",\"embeds\",\"media\"];var p=[],u=[],k=0;function j(){try{if(n.localStorage.getItem(y)===\"1\"){return true}}catch(e){}try{return n.sessionStorage.getItem(y)===\"1\"}catch(e){return false}}function S(){var e=n.nexusConsent;if(!e){return null}return typeof e.allows===\"function\"?e:null}function z(){var e=n.nexusConsent;if(!e){return false}if(typeof e.granted!==\"function\"){return false}for(var t=0;t<I.length;t++){try{if(e.granted(I[t])){return true}}catch(r){}}return false}function d(e){var t=S();if(t){try{return t.allows(e?e:\"maps\")===true}catch(r){return false}}return j()?true:z()}function D(e){try{var t=e?typeof e===\"string\"?e:e.url?e.url:e.href?e.href:String(e):\"\";var r=new URL(t,l.baseURI);if(r.protocol!==\"http:\"){if(r.protocol!==\"https:\"){return\"\"}}return r.origin===n.location.origin?\"\":r.hostname}catch(o){return\"\"}}function X(e){for(var t=0;t<C.length;t++){if(C[t].test(e)){return\"globe\"}}return\"maps\"}function f(e){var t=D(e);if(!t){return false}var r=false;for(var o=0;o<E.length;o++){if(E[o].test(t)){r=true;break}}if(!r){return false}if(d(X(t))){return false}k++;return true}var w=0;function c(){if(w){return}w=setTimeout(function(){w=0;L()},250)}function v(e){setTimeout(function(){try{e.dispatchEvent(new Event(\"error\"))}catch(t){}},0)}var N=n.fetch;if(N){n.fetch=function(e){if(f(e)){c();return Promise.reject(new TypeError(\"Blocked until the live map is loaded\"))}return N.apply(this,arguments)}}var g=n.XMLHttpRequest?n.XMLHttpRequest.prototype:null;if(g){var K=g.open,_=g.send;g.open=function(e,t){this.__nxLiveUrl=t;return K.apply(this,arguments)};g.send=function(){if(f(this.__nxLiveUrl)){c();v(this);return}return _.apply(this,arguments)}}var h=n.HTMLImageElement?Object.getOwnPropertyDescriptor(n.HTMLImageElement.prototype,\"src\"):null;if(h){if(h.set){Object.defineProperty(n.HTMLImageElement.prototype,\"src\",{configurable:true,enumerable:h.enumerable,get:h.get,set:function(e){if(f(e)){c();v(this);return}h.set.call(this,e)}})}}var O=n.Element?n.Element.prototype.setAttribute:null;if(O){n.Element.prototype.setAttribute=function(e,t){if(e===\"src\"){if(this.tagName===\"IMG\"){if(f(t)){c();v(this);return}}}return O.apply(this,arguments)}}function M(e){if(!e){return false}if(e.tagName===\"SCRIPT\"){return f(e.src)}if(e.tagName===\"LINK\"){return f(e.href)}return false}if(n.Node){var G=n.Node.prototype.appendChild,Y=n.Node.prototype.insertBefore;n.Node.prototype.appendChild=function(e){if(M(e)){c();v(e);return e}return G.apply(this,arguments)};n.Node.prototype.insertBefore=function(e){if(M(e)){c();v(e);return e}return Y.apply(this,arguments)}}var R='[data-nx-map=\"ctx-field\"],[data-nx-le],[data-nx-eo],[data-grx-globe],[data-grx-obs],.nx-sb[data-nx-sb]';function T(){var e=[l];var t=l.querySelectorAll(\"*\");for(var r=0;r<t.length;r++){if(t[r].shadowRoot){e.push(t[r].shadowRoot)}}return e}function B(){var e=S();if(e){if(typeof e.enable===\"function\"){try{e.enable(\"maps\",\"site\")}catch(t){}try{e.enable(\"globe\",\"site\")}catch(t){}m();return}}try{n.localStorage.setItem(y,\"1\")}catch(t){try{n.sessionStorage.setItem(y,\"1\")}catch(r){}}m()}function F(e,t){var r=e.createElement(\"button\");r.type=\"button\";r.textContent=A.label;r.setAttribute(\"style\",\"pointer-events:auto;min-height:44px;min-width:44px;padding:0 18px;cursor:pointer;font:inherit;font-weight:600;font-size:14px;line-height:1.2;border-radius:2px;border:1px solid var(--nx-border-strong,#8a8a8a);background:var(--nx-text-primary,#111);color:var(--nx-bg-canvas,#fff)\");var o=\"nx-live-note-\"+Math.floor(Math.random()*1e9);r.setAttribute(\"aria-describedby\",o);r.addEventListener(\"click\",B);r.addEventListener(\"focus\",function(){r.style.outline=\"2px solid var(--nx-focus,#4c9aff)\";r.style.outlineOffset=\"2px\"});r.addEventListener(\"blur\",function(){r.style.outline=\"\"});var a=e.createElement(\"span\");a.id=o;a.textContent=A.note;a.setAttribute(\"style\",\"pointer-events:none;font-size:12px;line-height:1.4;padding:3px 7px;border-radius:2px;background:var(--nx-bg-canvas,#fff);color:var(--nx-text-secondary,#333);border:1px solid var(--nx-border-hairline,#bbb);max-width:34em\");t.appendChild(r);t.appendChild(a)}function W(e){var t=e.ownerDocument;var r=t.createElement(\"div\");r.className=\"nx-live-gate\";r.setAttribute(\"style\",\"position:absolute;inset-block-start:10px;inset-inline-start:10px;z-index:6;display:flex;flex-direction:column;align-items:flex-start;gap:6px;max-width:calc(100% - 20px);pointer-events:none;font:inherit\");var o=S();var a=false;if(o){if(typeof o.placeholder===\"function\"){try{var s=t.createElement(\"div\");s.setAttribute(\"style\",\"pointer-events:auto\");r.appendChild(s);o.placeholder(s,\"maps\");a=s.childNodes.length>0;if(!a){r.removeChild(s)}}catch(i){a=false}}}if(!a){F(t,r)}return r}function L(){if(d(\"maps\")){return}var e=T();for(var t=0;t<e.length;t++){var r=e[t].querySelectorAll(R);for(var o=0;o<r.length;o++){var a=r[o];if(a.getAttribute(\"data-nx-live-gate\")){continue}if(a.closest(\".nx-live-gate\")){continue}var s=a.getBoundingClientRect();if(s.width<120){continue}var i=a.querySelector(\".le-stage,.eo-stage\");if(!i){if(a.hasAttribute(\"data-nx-le\")||a.hasAttribute(\"data-nx-eo\")){continue}i=a}a.setAttribute(\"data-nx-live-gate\",\"1\");var H=n.getComputedStyle(i);if(H.position===\"static\"){i.style.position=\"relative\"}var U=W(i);i.appendChild(U);u.push({host:a,wrap:U,stage:i})}}J()}function J(){for(var e=0;e<u.length;e++){var t=u[e].stage,r=t?t.querySelector(\".nx-wm-presets\"):null;var o=r?r.getBoundingClientRect():null;if(o?o.height>0:false){var a=Math.round(o.bottom-t.getBoundingClientRect().top+8)+\"px\";if(u[e].wrap.style.insetBlockStart!==a){u[e].wrap.style.insetBlockStart=a}}}}function Q(){for(var e=0;e<u.length;e++){try{u[e].wrap.remove()}catch(t){}try{u[e].host.removeAttribute(\"data-nx-live-gate\")}catch(t){}}u=[]}function V(e,t,r){p.push({name:e,selector:t,fn:r});if(x){try{r(\"on\")}catch(o){}}}function Z(e){for(var t=0;t<p.length;t++){try{p[t].fn(e)}catch(r){}}}function ee(){var e=[],t=T();for(var r=0;r<t.length;r++){var o=t[r].querySelectorAll(R);for(var a=0;a<o.length;a++){var s=false;for(var i=0;i<p.length;i++){try{if(o[a].matches(p[i].selector)){s=true;break}}catch(H){}}if(!s){e.push(o[a])}}}return e}function te(e){if(k===0){return}if(ee().length===0){return}var t=e?\"on\":\"off\";try{if(n.sessionStorage.getItem($)===t){return}n.sessionStorage.setItem($,t)}catch(r){return}n.location.reload()}var x=d(\"maps\");function m(){var e=d(\"maps\");if(e===x){if(!e){c()}return}x=e;if(e){Q()}Z(e?\"on\":\"off\");if(!e){c()}try{l.dispatchEvent(new CustomEvent(\"nexus:live-map\",{detail:{granted:e}}))}catch(t){}te(e)}function q(){L();if(n.MutationObserver){var e=new n.MutationObserver(function(){if(x){return}c()});e.observe(l.documentElement,{childList:true,subtree:true});setTimeout(function(){e.disconnect()},2e4)}}if(l.readyState===\"loading\"){l.addEventListener(\"DOMContentLoaded\",q)}else{q()}n.addEventListener(\"load\",function(){setTimeout(L,400)});l.addEventListener(\"nexus-consent:ready\",m);l.addEventListener(\"nexus-consent:change\",m);var P=0;var re=setInterval(function(){P++;m();if(P>40){clearInterval(re)}},400);n.NXLive={allowed:d,blocked:f,grant:B,register:V,hosts:E.length}})();<\/script><style>.pix-dark-mode-switch, .bx-color-mode-toggle, .color-switcher, .page-headline { display: none !important; }<\/style><script data-digest=\"1HDcdXHFCxvm20tlR9FHTN8XvrJhNFlbW03tKxwWJ8I=\" type=\"text\/x-nexus\" class=\"_iub_cs_skip\" id=\"nxe-es-load\" data-nx-run=\"estates\">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<\/script><script type=\"application\/json\" class=\"_iub_cs_skip\" id=\"nxe-assets\">{\"nxe-es-code\":{\"f\":\"es-code.d470dc7571c5.txt\",\"s\":\"sha256-1HDcdXHFCxvm20tlR9FHTN8XvrJhNFlbW03tKxwWJ8I=\",\"a\":\"\"},\"nxe-gl-base-od\":{\"f\":\"gl-base-od.e0622a96e899.txt\",\"s\":\"sha256-4GIqluiZJntyAGfMYGNoAXwwxMKhdab+7YEFBoLlZ5o=\",\"a\":\" data-nx-pack=\\\"deflate\\\"\"},\"nxe-map-context\":{\"f\":\"map-context.a965226720a1.txt\",\"s\":\"sha256-qWUiZyChbKiTtLWpFCZROVpB5QISXNBKfepOJ+zOkWA=\",\"a\":\" data-nx-pack=\\\"deflate\\\"\"},\"nxe-map-pool\":{\"f\":\"map-pool.c31dae84875a.txt\",\"s\":\"sha256-wx2uhIda8fEz4BK1sUQO7edPcs22Bvn4FE0NpYLDyEo=\",\"a\":\" data-nx-pack=\\\"deflate\\\"\"},\"nxe-ml-src-code\":{\"f\":\"ml-src-code.5a99eb24c979.txt\",\"s\":\"sha256-WpnrJMl5+Rt6GCPchYzz9V8Fw\/mt0BVdtedF3t\/o0+o=\",\"a\":\"\"},\"nxe-xt-code\":{\"f\":\"xt-code.371023489e89.txt\",\"s\":\"sha256-NxAjSJ6JoGGZsXu22d+kvA4nR9AJ6Njy+\/Xy2FIyELw=\",\"a\":\"\"}}<\/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\">if(typeof window.__nxStill!==\"function\")window.__nxStill=function(){try{if(document.documentElement.hasAttribute(\"data-nx-still\"))return true;if(document.body?\/(?:^|\\s)mdp-readabler-stop-animations(?:\\s|$)\/.test(document.body.className):false)return true;var g=JSON.parse(window.localStorage.getItem(\"readabler\")||\"{}\");return g?g.stopAnimations===true||g.stopAnimations===\"true\"||g.stopAnimations===1:false}catch(G){return false}};(function(){function g(e,a){var f=atob(e),s=new Uint8Array(f.length),r;for(r=0;r< f.length;r++){s[r]=f.charCodeAt(r)}var o=0,i=0,u=0,n=new Uint8Array(Math.max(1024,s.length*6)),t=0;var c=[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 d=[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 m=[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 y=[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 ct=[16,17,18,0,8,7,9,6,10,5,11,4,12,3,13,2,14,1,15];function b(w){while(u< w){if(o>=s.length){throw new Error(\"nx-inflate: truncated\")}i|=s[o++]<< u;u+=8}var h=i%(1<< w);i>>>=w;u-=w;return h}function S(w,h,A){var v=new Uint16Array(16),_=new Uint16Array(16),p=new Uint16Array(A),x;for(x=0;x< A;x++){v[w[h+x]]++}v[0]=0;for(x=1;x<16;x++){_[x]=_[x-1]+v[x-1]}for(x=0;x< A;x++){if(w[h+x]){p[_[w[h+x]]++]=x}}return{count:v,sym:p}}function X(w){var h=0,A=0,v=0,_,p;for(_=1;_<16;_++){h|=b(1);p=w.count[_];if(h-p< A){return w.sym[v+(h-A)]}v+=p;A+=p;A<<=1;h<<=1}throw new Error(\"nx-inflate: bad code\")}function B(w){if(t+w<=n.length){return}var h=new Uint8Array(Math.max(n.length*2,t+w));h.set(n.subarray(0,t));n=h}var j=null,W=null,Y=0;while(!Y){Y=b(1);var q=b(2),J,H;if(q===0){i=0;u=0;var D=s[o]|s[o+1]<<8;o+=4;B(D);n.set(s.subarray(o,o+D),t);t+=D;o+=D;continue}if(q===1){if(!j){var Z=new Uint8Array(288),k=new Uint8Array(30);for(r=0;r<288;r++){Z[r]=r<144?8:r<256?9:r<280?7:8}for(r=0;r<30;r++){k[r]=5}j=S(Z,0,288);W=S(k,0,30)}J=j;H=W}else if(q===2){var F=b(5)+257,tt=b(5)+1,dt=b(4)+4,N=new Uint8Array(320),nt=new Uint8Array(19);for(r=0;r< dt;r++){nt[ct[r]]=b(3)}var lt=S(nt,0,19),U=0;while(U< F+tt){var O=X(lt),R=0,et=0;if(O<16){N[U++]=O;continue}if(O===16){if(!U){throw new Error(\"nx-inflate: bad repeat\")}et=N[U-1];R=3+b(2)}else if(O===17){R=3+b(3)}else{R=11+b(7)}while(R--){N[U++]=et}}J=S(N,0,F);H=S(N,F,tt)}else{throw new Error(\"nx-inflate: bad block\")}for(;;){var E=X(J);if(E<256){B(1);n[t++]=E;continue}if(E===256){break}E-=257;var rt=c[E]+b(d[E]),at=X(H),it=m[at]+b(y[at]);if(it>t){throw new Error(\"nx-inflate: bad distance\")}B(rt);for(var ot=0;ot< rt;ot++){n[t]=n[t-it];t++}}}return a?n.subarray(0,t):new TextDecoder(\"utf-8\").decode(n.subarray(0,t))}if(typeof window.__nxInflate!==\"function\"){window.__nxInflate=g}function G(e){var a=g(e,true),f=a[0]|a[1]<<8|a[2]<<16|a[3]<<24;var s=a.subarray(4+f);function r(o,i){var u=\"\",n=s.subarray(o,o+i);for(var t=0;i>t;t+=8192){u+=String.fromCharCode.apply(null,n.subarray(t,t+8192))}return btoa(u)}return JSON.parse(new TextDecoder(\"utf-8\").decode(a.subarray(4,4+f)),function(o,i){return i?i.$b?r(i.$b[0],i.$b[1]):i:i})}if(typeof window.__nxUnpack!==\"function\"){window.__nxUnpack=G}if(typeof window.__nxLate!==\"function\"){window.__nxLate=function(e,a){if(document.readyState!==\"loading\"){return null}if(!document.querySelector('template[data-nx-libs-last~=\"'+e+'\"]')){return null}return new Promise(function(f,s){document.addEventListener(\"DOMContentLoaded\",function(){var r=a();if(r){f(r)}else{s(new Error(\"nx-late: \"+e))}},{once:true})})}}if(typeof window.__nxAsset!==\"function\"){var I={};window.__nxAsset=function(e,a){if(I[e]){return I[e]}var f=document.getElementById(\"nxe-boot\"),s=[],r=window.NX_ASSET_BASE,o=f?f.getAttribute(\"data-nx-asset-base\"):\"\";if(r){s.push(String(r))}if(o){s.push(o)}s.push(\"\/wp-content\/uploads\/nexus-assets\/\");function i(t){if(!a){return t}return crypto.subtle.digest(\"SHA-256\",new TextEncoder().encode(t)).then(function(c){if(\"sha256-\"+btoa(String.fromCharCode.apply(null,new Uint8Array(c)))!==a){throw new Error(\"nx-asset: integrity\")}return t})}function u(t){if(t>=s.length){return Promise.reject(new Error(\"nx-asset: \"+e))}var c=s[t].charAt(s[t].length-1)===\"\/\"?s[t]:s[t]+\"\/\";return fetch(c+e,{credentials:\"omit\"}).then(function(d){if(!d.ok){throw new Error(\"nx-asset: \"+d.status)}return d.text()}).then(i).then(null,function(){return u(t+1)})}var n=u(0);I[e]=n;n.then(null,function(){delete I[e]});return n}}if(typeof window.__nxText!==\"function\"){window.__nxText=function(e){if(!e){return Promise.reject(new Error(\"nx-text: none\"))}var a=e.getAttribute(\"data-nx-asset\");if(a?!e.textContent:false){return window.__nxAsset(a,e.getAttribute(\"data-nx-sri\")||\"\")}return Promise.resolve(e.textContent)}}function ut(e){var a=e.getAttribute(\"data-nx-asset\");if(a?!e.textContent:false){return window.__nxAsset(a,e.getAttribute(\"data-nx-sri\")||\"\").then(function(n){return window.__nxInflate(n)})}var f=e.getAttribute(\"data-nx-src\");if(f?!e.textContent:false){var s=window.NX_RUNTIME||{},r=s.base||window.NX_RUNTIME_BASE||\"\",o=(s.files||{})[f.replace(\/\\.[0-9a-f]{12}(\\.js)$\/,\"$1\")]||null,i=[];if(r){i.push([r+f,e.getAttribute(\"data-nx-sri\")||\"\"])}if(o?!!o.url:false){i.push([o.url,o.sri||\"\"])}var u=function(n){if(n>=i.length){return Promise.reject(new Error(\"nx-lazy: \"+f))}var t={credentials:\"same-origin\"};if(i[n][1]){t.integrity=i[n][1]}return fetch(i[n][0],t).then(function(c){if(!c.ok){throw new Error(\"nx-lazy: \"+c.status)}return c.text()}).then(null,function(){return u(n+1)})};return u(0)}return Promise.resolve(window.__nxInflate(e.textContent))}function st(e){var a=e[0].getAttribute(\"data-nx-on-sight\")||\"[data-nx-map]\",f=null,s=false;function r(){if(s){return}s=true;if(f){f.disconnect()}var i=document.querySelectorAll(\"script[data-nx-on-sight][data-nx-asset]:not([data-nx-run])\"),u=[];for(var n=0;n!==i.length;n++){(function(t){if(t.textContent){return}u.push(window.__nxAsset(t.getAttribute(\"data-nx-asset\"),t.getAttribute(\"data-nx-sri\")||\"\").then(function(c){t.textContent=c;t.removeAttribute(\"data-nx-asset\")},function(){t.setAttribute(\"data-nx-ran\",\"failed\")}))})(i[n])}Promise.all(u.concat(e.map(function(t){return ut(t).then(null,function(){return null})}))).then(function(t){var c=t.slice(u.length);for(var d=0;d!==e.length;d++){if(c[d]===null){e[d].setAttribute(\"data-nx-ran\",\"failed\");continue}var m=document.createElement(\"script\");m.className=\"_iub_cs_skip\";m.setAttribute(\"data-nx-runtime\",e[d].getAttribute(\"data-nx-run\"));m.textContent=c[d];(document.head||document.documentElement).appendChild(m);e[d].setAttribute(\"data-nx-ran\",\"true\")}try{document.dispatchEvent(new CustomEvent(\"nexus:lazy-ready\"))}catch(y){}})}function o(){if(s){return}var i=[document],u=document.querySelectorAll(\".nxe-host\"),n=[],t,c;for(t=0;t!==u.length;t++){if(u[t].shadowRoot){i.push(u[t].shadowRoot)}}for(t=0;t!==i.length;t++){c=i[t].querySelectorAll(a);for(var d=0;d!==c.length;d++){n.push(c[d])}}if(!n.length){return}if(!window.IntersectionObserver){r();return}f=f||new IntersectionObserver(function(m){for(var y=0;y!==m.length;y++){if(m[y].isIntersecting){r();return}}},{rootMargin:\"800px 0px 800px 0px\"});for(t=0;t!==n.length;t++){if(!n[t].__nxLazySeen){n[t].__nxLazySeen=true;f.observe(n[t])}}}document.addEventListener(\"DOMContentLoaded\",o);document.addEventListener(\"nexus:runtime-ready\",o);window.addEventListener(\"load\",o);setTimeout(o,0)}var K=document.querySelectorAll(\"script[data-nx-run]\"),C=[],$=[];for(var z=0;z< K.length;z++){var l=K[z];if(l.hasAttribute(\"data-nx-ran\")){continue}if(l.hasAttribute(\"data-nx-on-sight\")){l.setAttribute(\"data-nx-ran\",\"waiting\");$.push(l);continue}if(l.hasAttribute(\"data-nx-src\")?!l.textContent:false){l.setAttribute(\"data-nx-ran\",\"loading\");C.push(l);continue}if(l.hasAttribute(\"data-nx-asset\")?!l.textContent:false){l.setAttribute(\"data-nx-ran\",\"loading\");C.push(l);continue}l.setAttribute(\"data-nx-ran\",\"true\");try{var L=document.createElement(\"script\");L.className=\"_iub_cs_skip\";L.setAttribute(\"data-nx-runtime\",l.getAttribute(\"data-nx-run\"));L.textContent=window.__nxInflate(l.textContent);(document.head||document.documentElement).appendChild(L)}catch(e){l.setAttribute(\"data-nx-ran\",\"failed\")}}if($.length){st($)}if(!C.length){return}var Q=window.NX_RUNTIME||{},V=Q.base||window.NX_RUNTIME_BASE||\"\",T=window.__nxMount,P=[];if(typeof T===\"function\"){window.__nxMount=function(e){P.push(e)}}function ft(){if(typeof T===\"function\"){window.__nxMount=T;for(var e=0;e< P.length;e++){try{T(P[e])}catch(a){}}}try{document.dispatchEvent(new CustomEvent(\"nexus:runtime-ready\"))}catch(a){}}function M(e){if(e>=C.length){ft();return}var a=C[e],f=a.getAttribute(\"data-nx-src\")||\"\",s=Q.files||{},r=[];if(a.hasAttribute(\"data-nx-asset\")){window.__nxAsset(a.getAttribute(\"data-nx-asset\"),a.getAttribute(\"data-nx-sri\")||\"\").then(function(u){var n=document.createElement(\"script\");n.className=\"_iub_cs_skip\";n.setAttribute(\"data-nx-runtime\",a.getAttribute(\"data-nx-run\"));n.textContent=window.__nxInflate(u);(document.head||document.documentElement).appendChild(n);a.setAttribute(\"data-nx-ran\",\"true\")}).then(null,function(){a.setAttribute(\"data-nx-ran\",\"failed\")}).then(function(){M(e+1)});return}var o=s[f.replace(\/\\.[0-9a-f]{12}(\\.js)$\/,\"$1\")]||null;if(V){r.push([V+f,a.getAttribute(\"data-nx-sri\")||\"\"])}if(o?!!o.url:false){r.push([o.url,o.sri||\"\"])}function i(u){if(u>=r.length){a.setAttribute(\"data-nx-ran\",\"failed\");M(e+1);return}var n=document.createElement(\"script\");n.className=\"_iub_cs_skip\";n.setAttribute(\"data-nx-runtime\",a.getAttribute(\"data-nx-run\"));if(r[u][1]){n.integrity=r[u][1]}n.onload=function(){a.setAttribute(\"data-nx-ran\",\"true\");M(e+1)};n.onerror=function(){if(n.parentNode){n.parentNode.removeChild(n)}i(u+1)};n.src=r[u][0];(document.head||document.documentElement).appendChild(n)}i(0)}M(0)})();<\/script><div class=\"nxe-host\" data-nxe-embed=\"n-g-p08-stem-ai-ml\"><div class=\"nxe\" id=\"nxe-nnzvjhlu\" data-inst=\"gcri\" data-theme=\"dark\" data-grammar=\"atlas\" data-domain=\"project\" data-arche=\"method\" data-density=\"compact\" lang=\"en\" data-nx-css=\"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\"><style>#nxe-nnzvjhlu.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-nnzvjhlu.nxe :where(*,*::before,*::after){box-sizing:inherit}#nxe-nnzvjhlu.nxe :where(section,header,footer,nav,aside){display:block;margin:0 0 32px}#nxe-nnzvjhlu.nxe :where(h1,h2,h3,h4){line-height:1.2;margin:24px 0 12px;font-weight:600}#nxe-nnzvjhlu.nxe :where(p,ul,ol,dl,table,figure,details){margin:0 0 16px}#nxe-nnzvjhlu.nxe :where(ul,ol){padding-left:20px}#nxe-nnzvjhlu.nxe a{color:inherit;text-decoration:underline;text-underline-offset:2px}#nxe-nnzvjhlu.nxe table{border-collapse:collapse;width:100%;display:block;overflow-x:auto}#nxe-nnzvjhlu.nxe :where(th,td){padding:6px 8px;text-align:left;vertical-align:top;border-bottom:1px solid;border-color:inherit}#nxe-nnzvjhlu.nxe svg{display:none}#nxe-nnzvjhlu.nxe svg.nx-i{display:inline-block;width:18px;height:18px;vertical-align:-3px}#nxe-nnzvjhlu.nxe :where([role=\"tablist\"],[role=\"toolbar\"],[hidden],button,.nx-theme-toggle,.nx-live-slot){display:none}#nxe-nnzvjhlu.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-scene nx-es-scene\" data-scene=\"text\" data-ground=\"none\" data-es-org=\"gcri\" aria-labelledby=\"nx-scene-title\"><div class=\"nx-scene-intro nx-es-intro\"><p class=\"nx-eyebrow\"><abbr title=\"The Global Centre for Risk and Innovation\">GCRI<\/abbr> \u00b7 Artificial Intelligence and Autonomous Systems<\/p><span class=\"nx-scene-rule\" aria-hidden=\"true\"><\/span><h1 id=\"nx-scene-title\" class=\"nx-scene-title\">Machine learning behaviour<\/h1><p class=\"nx-scene-lede\">Artificial intelligence and machine learning is the technology this platform stewards for the network. Here it is studied as a technology whose behaviour must be understood: how models learn from data, generalise, fail outside their training distribution and are adapted to specific uses.<\/p><div class=\"nx-actions\"><a class=\"nx-btn\" data-variant=\"primary\" href=\"https:\/\/therisk.global\/ai-autonomous-systems\/write-to-us\/contact\/\">Make an enquiry <svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><a class=\"nx-btn\" data-variant=\"secondary\" href=\"https:\/\/therisk.global\/ai-autonomous-systems\/stem\/\">Back to STEM foundation of Artificial Intelligence and Autonomous Systems<\/a><\/div><\/div><\/section><section id=\"nx-s-1-sec\" class=\"nx-section\" data-hue=\"project\" data-tone=\"alt\" aria-labelledby=\"nx-s-1\"><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\">Register<\/span><\/div><h2 id=\"nx-s-1\" class=\"nx-section-title\">Machine learning behaviour on the record.<\/h2><\/div><div class=\"nx-section-body\"><div class=\"nx-es-fig\" data-es-layout=\"console\"><div class=\"nx-hero-panel\"><div class=\"nx-register\"><div class=\"nx-register-head\"><span class=\"nx-eyebrow\"><strong>Method record<\/strong><\/span><span class=\"nx-eyebrow\">STEM method<\/span><\/div><ol class=\"nx-register-rows\"><li><div><span class=\"nx-register-k\">Method<\/span><span class=\"nx-register-v\">Artificial intelligence and machine learning<\/span><\/div><\/li><li><div><span class=\"nx-register-k\">Steward<\/span><span class=\"nx-register-v\">Artificial Intelligence and Autonomous <a class=\"nx-link nx-term\" href=\"https:\/\/docs.therisk.global\/organization\/systems\" data-tc=\"Systems\" data-tk=\"Nexus Standards\" data-td=\"The twenty Nexus systems, NXS-CMD to NXS-LAB\">Systems<\/a><\/span><\/div><\/li><li><div><span class=\"nx-register-k\">Applied in<\/span><span class=\"nx-register-v\">Artificial Intelligence and Autonomous Systems<\/span><\/div><\/li><li><div><span class=\"nx-register-k\">Data<\/span><span class=\"nx-register-v\"><a class=\"nx-link\" href=\"#nx-sec-data\">3 inputs<\/a><\/span><\/div><\/li><li><div><span class=\"nx-register-k\">Checks<\/span><span class=\"nx-register-v\"><a class=\"nx-link\" href=\"#nx-sec-checks\">3 validation steps<\/a><\/span><\/div><\/li><li><div><span class=\"nx-register-k\">Canonical page<\/span><span class=\"nx-register-v\"><a class=\"nx-link\" href=\"https:\/\/therisk.global\/stem\/technology\/ai-ml\/\">STEM Commons<\/a><\/span><\/div><\/li><\/ol><\/div><\/div><\/div><\/div><\/div><\/section><section class=\"nx-section\" data-band=\"auto\" data-band-id=\"nx-band-1\" data-hue=\"infrastructure\" aria-label=\"Application, Simulator, Data, Methods and evidence\"><div class=\"nx-container\"><div class=\"nx-hm-band nx-band\" data-nx-tabs=\"true\" data-active=\"p1\"><div class=\"nx-tablist nx-hm-tabs\" role=\"tablist\" aria-label=\"Application, Simulator, Data, Methods and evidence\"><button class=\"nx-tab\" type=\"button\" role=\"tab\" id=\"nx-band-1-tab-1\" aria-controls=\"nx-sec-application\" aria-selected=\"true\" data-key=\"p1\">Application<\/button><button class=\"nx-tab\" type=\"button\" role=\"tab\" id=\"nx-band-1-tab-2\" aria-controls=\"nx-sec-instrument\" aria-selected=\"false\" data-key=\"p2\">Simulator<\/button><button class=\"nx-tab\" type=\"button\" role=\"tab\" id=\"nx-band-1-tab-3\" aria-controls=\"nx-sec-data\" aria-selected=\"false\" data-key=\"p3\">Data<\/button><button class=\"nx-tab\" type=\"button\" role=\"tab\" id=\"nx-band-1-tab-4\" aria-controls=\"nx-sec-audience\" aria-selected=\"false\" data-key=\"p4\">Methods and evidence<\/button><\/div><section id=\"nx-sec-application\" class=\"nx-section nx-band-panel\" data-hue=\"infrastructure\" data-variant=\"split\" aria-labelledby=\"nx-s-2\" role=\"tabpanel\" data-band-tab=\"nx-band-1-tab-1\"><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\">Application<\/span><\/div><h2 id=\"nx-s-2\" class=\"nx-section-title\">What artificial intelligence and machine learning does here.<\/h2><p class=\"nx-lede\">Artificial intelligence and machine learning develop algorithms that learn patterns from data to classify, predict or generate outputs. They are increasingly used for hazard detection, forecasting and decision support, and are themselves a source of risk. In artificial intelligence and data and compute, machine learning can scale detection and forecasting across interconnected <a class=\"nx-link nx-term\" href=\"https:\/\/docs.therisk.global\/organization\/systems\" data-tc=\"Systems\" data-tk=\"Nexus Standards\" data-td=\"The twenty Nexus systems, NXS-CMD to NXS-LAB\">systems<\/a>. Its own failure modes must be weighed when automated outputs inform consequential decisions. The canonical description is on <a class=\"nx-link\" href=\"https:\/\/therisk.global\/stem\/technology\/ai-ml\/\">the STEM Commons<\/a>.<\/p><\/div><div class=\"nx-section-body\"><div class=\"nx-cols\"><article><div class=\"nx-glyph-row\"><span class=\"nx-glyph\" data-tone=\"accent\" data-size=\"lg\"><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 5a3 3 0 1 0-5.997.125 4 4 0 0 0-2.526 5.77 4 4 0 0 0 .556 6.588A4 4 0 1 0 12 18Z\"\/><path d=\"M9 13a4.5 4.5 0 0 0 3-4\"\/><path d=\"M6.003 5.125A3 3 0 0 0 6.401 6.5\"\/><path d=\"M3.477 10.896a4 4 0 0 1 .585-.396\"\/><path d=\"M6 18a4 4 0 0 1-1.967-.516\"\/><path d=\"M12 13h4\"\/><path d=\"M12 18h6a2 2 0 0 1 2 2v1\"\/><path d=\"M12 8h8\"\/><path d=\"M16 8V5a2 2 0 0 1 2-2\"\/><circle cx=\"16\" cy=\"13\" r=\".5\"\/><circle cx=\"18\" cy=\"3\" r=\".5\"\/><circle cx=\"20\" cy=\"21\" r=\".5\"\/><circle cx=\"20\" cy=\"8\" r=\".5\"\/><\/svg><\/span><\/div><h3 class=\"nx-h3 nx-mt-2\">Distribution shift analysis<\/h3><p class=\"nx-p nx-mt-1\">Model performance is compared between training data and deployment data to detect shifts in inputs or labels that reduce accuracy, with monitoring set up to catch drift.<\/p><\/article><article><div class=\"nx-glyph-row\"><span class=\"nx-glyph\" data-tone=\"accent\" data-size=\"lg\"><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><\/div><h3 class=\"nx-h3 nx-mt-2\">Fine-tuning for a specific use<\/h3><p class=\"nx-p nx-mt-1\">General models are adapted to a narrow task with curated data, and the adapted model is evaluated for both task performance and retained safety behaviour.<\/p><\/article><article><div class=\"nx-glyph-row\"><span class=\"nx-glyph\" data-tone=\"accent\" data-size=\"lg\"><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=\"M2 12s3-7 10-7 10 7 10 7-3 7-10 7-10-7-10-7Z\"\/><circle cx=\"12\" cy=\"12\" r=\"3\"\/><\/svg><\/span><\/div><h3 class=\"nx-h3 nx-mt-2\">Interpretability methods<\/h3><p class=\"nx-p nx-mt-1\">Feature attribution and probing methods are used to examine which inputs drive a model&#8217;s outputs, with their limits as explanations stated.<\/p><\/article><\/div><\/div><\/div><\/section><section id=\"nx-sec-instrument\" class=\"nx-section nx-band-panel\" data-hue=\"project\" data-ns-layout=\"es-wide\" aria-labelledby=\"nx-s-3\" role=\"tabpanel\" data-band-tab=\"nx-band-1-tab-2\"><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\">Simulator<\/span><\/div><h2 id=\"nx-s-3\" class=\"nx-section-title\">A threshold model of collective behaviour.<\/h2><p class=\"nx-lede\">Each agent acts once enough of its contacts have; small changes in the mean threshold or the early movers decide whether the behaviour stays local or cascades through the population. Teaching model with stated illustrative values, not a prediction.<\/p><\/div><div class=\"nx-section-body\"><figure class=\"nx-in nx-fk\" id=\"nx-in-ai-autonomous-systems-stem-ai-ml-0\" data-nxi=\"abm\" data-nxi-id=\"abm\" data-nxi-z=\"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\" aria-label=\"A threshold model of collective behaviour\"><div class=\"nx-in-body\"><div class=\"nx-in-main\"><div class=\"nx-in-tabs\" data-nxi-tabs=\"true\" aria-label=\"Views of the model\" hidden><button type=\"button\" class=\"nx-in-tab\" data-nxi-view=\"result\">Result<\/button><button type=\"button\" class=\"nx-in-tab\" data-nxi-view=\"agents\">Agents<\/button><button type=\"button\" class=\"nx-in-tab\" data-nxi-view=\"phase\">Tipping point<\/button><button type=\"button\" class=\"nx-in-tab\" data-nxi-view=\"tornado\">Sensitivity<\/button><button type=\"button\" class=\"nx-in-tab\" data-nxi-view=\"loops\">Loops<\/button><\/div><div class=\"nx-in-stage\" id=\"nx-in-ai-autonomous-systems-stem-ai-ml-0-stage\" data-nxi-stage=\"true\" data-view=\"result\"><svg class=\"nx-in-svg\" viewBox=\"0 0 720 340\" role=\"img\" aria-label=\"Adoption\" data-nx-z=\"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\"><\/svg><\/div><p class=\"nx-in-tip\" data-nxi-tip=\"true\" aria-live=\"polite\"><\/p><div class=\"nx-in-bar\" data-nxi-bar=\"true\" role=\"group\" aria-label=\"Play over time\" hidden><button type=\"button\" class=\"nx-in-btn\" data-nxi-prev=\"true\" aria-label=\"Step back\">&#8249;<\/button><button type=\"button\" class=\"nx-in-btn\" data-nxi-play=\"true\" aria-pressed=\"false\">Play<\/button><button type=\"button\" class=\"nx-in-btn\" data-nxi-next=\"true\" aria-label=\"Step forward\">&#8250;<\/button><span class=\"nx-in-step\" data-nxi-step=\"true\" aria-live=\"polite\"><\/span><\/div><\/div><div class=\"nx-in-side\"><p class=\"nx-eyebrow\">Controls<\/p><div class=\"nx-in-ctl\" data-nxi-ctl=\"true\"><dl class=\"nx-in-pv\"><div><dt>Mean threshold (share of neighbours)<\/dt><dd>18%<\/dd><\/div><div><dt>Spread of thresholds<\/dt><dd>8 points<\/dd><\/div><div><dt>Early movers<\/dt><dd>2.0% of people<\/dd><\/div><div><dt>Contacts per person<\/dt><dd>8<\/dd><\/div><\/dl><\/div><p class=\"nx-eyebrow\">Readouts<\/p><dl class=\"nx-in-out\" data-nxi-out=\"true\" aria-live=\"polite\"><div><dt>Share who acted in the end<\/dt><dd>100%<\/dd><\/div><div><dt>Global cascade<\/dt><dd>Yes<\/dd><\/div><div><dt>Steps to settle<\/dt><dd>10<\/dd><\/div><\/dl><\/div><\/div><details class=\"nx-in-more\"><summary>Numbers, methods and sources<\/summary><div class=\"nx-in-more-in\"><div class=\"nx-in-tblw\" data-nxi-table=\"true\"><table class=\"nx-in-tbl\"><thead><tr><th scope=\"col\">Step<\/th><th scope=\"col\">Share who have acted (%)<\/th><\/tr><\/thead><tbody><tr><th scope=\"row\">0<\/th><td>2.3<\/td><\/tr><tr><th scope=\"row\">1<\/th><td>6.5<\/td><\/tr><tr><th scope=\"row\">2<\/th><td>16.0<\/td><\/tr><tr><th scope=\"row\">3<\/th><td>28.5<\/td><\/tr><tr><th scope=\"row\">4<\/th><td>47.0<\/td><\/tr><tr><th scope=\"row\">5<\/th><td>65.3<\/td><\/tr><tr><th scope=\"row\">6<\/th><td>80.3<\/td><\/tr><tr><th scope=\"row\">7<\/th><td>93.0<\/td><\/tr><tr><th scope=\"row\">8<\/th><td>99.3<\/td><\/tr><tr><th scope=\"row\">9<\/th><td>100.0<\/td><\/tr><tr><th scope=\"row\">10<\/th><td>100.0<\/td><\/tr><\/tbody><\/table><\/div><button type=\"button\" class=\"nx-in-btn\" data-nxi-csv=\"true\" hidden>Download the numbers (CSV)<\/button><h4 class=\"nx-h4\">Methods<\/h4><ol class=\"nx-in-meth\"><li>Threshold models of collective behaviour <span class=\"nx-in-cite\">Granovetter (1978), American Journal of Sociology 83<\/span><\/li><li>Global cascades on random networks <span class=\"nx-in-cite\">Watts (2002), PNAS 99<\/span><\/li><li>Complex contagions <span class=\"nx-in-cite\">Centola and Macy (2007), American Journal of Sociology 113<\/span><\/li><li>One-at-a-time sensitivity (tornado) <span class=\"nx-in-cite\">Saltelli and others (2008), Global Sensitivity Analysis: The Primer, Wiley<\/span><\/li><\/ol><h4 class=\"nx-h4\">Data<\/h4><p class=\"nx-p\">No observed data enters this model: it runs on the teaching values set by the controls, to be replaced with a programme&#8217;s own evidence.<\/p><\/div><\/details><figcaption class=\"nx-caption\">A threshold model of collective behaviour. Illustrative teaching model: it runs on the values set by the controls, not on measured data. Methods: Threshold models of collective behaviour (Granovetter (1978), American Journal of Sociology 83); Global cascades on random networks (Watts (2002), PNAS 99); Complex contagions (Centola and Macy (2007), American Journal of Sociology 113).<\/figcaption><div class=\"nx-in-pls\"><p class=\"nx-eyebrow\">Across the six pillars, led by Nexus Consortium<\/p><ul class=\"nx-in-pl\" aria-label=\"Where this instrument connects across the pillars\"><li data-pillar=\"labs\" data-hue=\"social\"><span class=\"nx-pl-dot\" aria-hidden=\"true\"><\/span><span class=\"nx-in-plv\">Experiments<\/span> <a class=\"nx-link\" href=\"https:\/\/therisk.global\/nexus-labs\/\">Labs<\/a><\/li><li data-pillar=\"foundry\" data-hue=\"infrastructure\"><span class=\"nx-pl-dot\" aria-hidden=\"true\"><\/span><span class=\"nx-in-plv\">Built in<\/span> <a class=\"nx-link\" href=\"https:\/\/therisk.global\/ai-autonomous-systems\/foundry\/\">Artificial Intelligence and Autonomous Systems in Foundry<\/a><\/li><li data-pillar=\"universe\" data-hue=\"water\"><span class=\"nx-pl-dot\" aria-hidden=\"true\"><\/span><span class=\"nx-in-plv\">Tested at<\/span> <a class=\"nx-link\" href=\"https:\/\/therisk.global\/ai-autonomous-systems\/universe\/\">Artificial Intelligence and Autonomous Systems in Universe<\/a><\/li><li data-pillar=\"standards\" data-hue=\"biodiversity\"><span class=\"nx-pl-dot\" aria-hidden=\"true\"><\/span><span class=\"nx-in-plv\">Codified in<\/span> <a class=\"nx-link\" href=\"https:\/\/nexustandards.com\/digital-ai\/\">Digital, Cyber and AI in Standards<\/a><\/li><li data-pillar=\"reports\" data-hue=\"energy\"><span class=\"nx-pl-dot\" aria-hidden=\"true\"><\/span><span class=\"nx-in-plv\">Published in<\/span> <a class=\"nx-link\" href=\"https:\/\/therisk.global\/ai-autonomous-systems\/reports\/\">Artificial Intelligence and Autonomous Systems in Reports<\/a><\/li><li data-pillar=\"academy\" data-hue=\"risk\"><span class=\"nx-pl-dot\" aria-hidden=\"true\"><\/span><span class=\"nx-in-plv\">Taught in<\/span> <a class=\"nx-link\" href=\"https:\/\/risksacademy.com\/technology-digital\/\">Faculty of Technology and Digital Systems in Academy<\/a><\/li><\/ul><\/div><span class=\"nx-in-keep\" hidden><i data-c=\"c1\"><\/i><i data-c=\"c2\"><\/i><i data-c=\"c3\"><\/i><i data-c=\"c4\"><\/i><i data-c=\"c5\"><\/i><i data-c=\"c6\"><\/i><i data-c=\"l0\"><\/i><i data-c=\"l1\"><\/i><i data-c=\"l2\"><\/i><i data-c=\"l3\"><\/i><i data-c=\"l5\"><\/i><i data-c=\"h3\"><\/i><i data-c=\"neg\"><\/i><i data-c=\"on0\"><\/i><i data-c=\"on1\"><\/i><i data-l=\"0\"><\/i><i data-l=\"1\"><\/i><i data-l=\"2\"><\/i><i data-l=\"3\"><\/i><i data-l=\"4\"><\/i><i data-l=\"5\"><\/i><i data-on=\"0\"><\/i><i data-on=\"1\"><\/i><i data-pol=\"p\"><\/i><i data-pol=\"n\"><\/i><i data-k=\"R\"><\/i><i data-k=\"B\"><\/i><i data-k=\"dash\"><\/i><i data-dash=\"true\"><\/i><i data-view=\"result\"><\/i><i data-nxi-live=\"on\"><\/i><i data-nxi-live=\"off\"><\/i><i class=\"nx-in-svg nx-in-grid nx-in-ax nx-in-tk nx-in-at nx-in-ln nx-in-band nx-in-lbl nx-in-lk nx-in-ref nx-in-reft nx-in-mk nx-in-x nx-in-rect nx-in-val nx-in-rowl nx-in-cl nx-in-arrow nx-in-pol nx-in-cln nx-in-loop nx-in-edge nx-in-node nx-in-nl nx-in-reach nx-in-trib nx-in-agent nx-in-flow nx-in-legend nx-in-sw nx-in-note nx-in-loops nx-in-3d nx-in-cv nx-in-field nx-in-fl nx-in-fv nx-in-range nx-in-btn nx-in-err nx-in-tbl nx-in-cite\"><\/i><\/span><\/figure><\/div><\/div><\/section><section id=\"nx-sec-data\" class=\"nx-section nx-band-panel\" data-hue=\"infrastructure\" aria-labelledby=\"nx-s-4\" role=\"tabpanel\" data-band-tab=\"nx-band-1-tab-3\"><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\">Data<\/span><\/div><h2 id=\"nx-s-4\" class=\"nx-section-title\">What the method consumes.<\/h2><p class=\"nx-lede\">Datasets and records used by artificial intelligence and machine learning on this platform, with the role each plays and the label its values carry. Sources are listed in <a class=\"nx-link\" href=\"https:\/\/therisk.global\/ai-autonomous-systems\/research\/sources\/\">the source register<\/a>.<\/p><\/div><div class=\"nx-section-body\" data-lv=\"register\"><div class=\"nx-table-wrap\"><table class=\"nx-table\" data-stack=\"true\"><thead><tr><th scope=\"col\">Data<\/th><th scope=\"col\">Role<\/th><th scope=\"col\">Label<\/th><\/tr><\/thead><tbody><tr><td data-label=\"Data\">Training and validation datasets with documentation<\/td><td data-label=\"Role\">Model development<\/td><td data-label=\"Label\"><span class=\"nx-chip\">DECLARED<\/span><\/td><\/tr><tr><td data-label=\"Data\">Deployment input logs<\/td><td data-label=\"Role\">Shift detection<\/td><td data-label=\"Label\"><span class=\"nx-chip\">OBSERVED<\/span><\/td><\/tr><tr><td data-label=\"Data\">Benchmark suites<\/td><td data-label=\"Role\">Comparison<\/td><td data-label=\"Label\"><span class=\"nx-chip\">REPORTED<\/span><\/td><\/tr><\/tbody><\/table><\/div><div class=\"nx-live-slot\" data-nxf-slot=\"data\" data-nxf-kind=\"data\" data-nxf-resource=\"nexus:\/\/gcri\/G-P08\/stem\/ai-ml\/#data\" data-nxf-title=\"Method record\" data-nxf-height=\"520\" hidden><\/div><div class=\"nx-live-slot\" data-nxf-slot=\"evidence\" data-nxf-kind=\"evidence\" data-nxf-resource=\"nexus:\/\/gcri\/G-P08\/stem\/ai-ml\/#evidence\" data-nxf-title=\"Evidence\" data-nxf-height=\"520\" hidden><\/div><\/div><\/div><\/section><section id=\"nx-sec-audience\" class=\"nx-section nx-band-panel\" data-hue=\"project\" aria-labelledby=\"nx-s-5\" role=\"tabpanel\" data-band-tab=\"nx-band-1-tab-4\"><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\">Methods and evidence<\/span><\/div><h2 id=\"nx-s-5\" class=\"nx-section-title\">How machine learning behaviour is modelled, measured and tested.<\/h2><p class=\"nx-lede\">The model, its assumptions, its data and its uncertainty, in the form a scientist or engineer would check. Structures follow published methods. The values drawn are illustrative, not findings.<\/p><\/div><div class=\"nx-section-body\"><div class=\"nx-au\" data-au-voice=\"science\" data-au-n=\"1\"><figure class=\"nx-au-c\" data-au-fig=\"method-panel\"><h3 class=\"nx-h4\">Model, equations and assumptions<\/h3><p class=\"nx-p nx-au-what\">The model structure used for machine learning behaviour, the assumptions it makes and how each is tested.<\/p><div class=\"nx-au-eqs\"><div class=\"nx-au-eq\"><span class=\"nx-au-k\">Risk of element i<\/span><code>R<sub>i<\/sub> = f(H<sub>i<\/sub>, E<sub>i<\/sub>, V<sub>i<\/sub>, C<sub>i<\/sub>)<\/code><span class=\"nx-au-n\">hazard, exposure, vulnerability and coping capacity<\/span><\/div><div class=\"nx-au-eq\"><span class=\"nx-au-k\">Expected annual loss<\/span><code>EAL = &#8721;<sub>k<\/sub> p<sub>k<\/sub> &#183; L<sub>k<\/sub><\/code><span class=\"nx-au-n\">over events k with annual probability p<sub>k<\/sub><\/span><\/div><div class=\"nx-au-eq\"><span class=\"nx-au-k\">Coupled systems<\/span><code>x<sub>t+1<\/sub> = A x<sub>t<\/sub> + B u<sub>t<\/sub> + &#949;<sub>t<\/sub><\/code><span class=\"nx-au-n\">A carries the couplings, u the shock, &#949; the error<\/span><\/div><\/div><div class=\"nx-table-wrap\"><table class=\"nx-table\" data-stack=\"true\"><thead><tr><th scope=\"col\">Assumption<\/th><th scope=\"col\">Why it is made<\/th><th scope=\"col\">How it is tested<\/th><\/tr><\/thead><tbody><tr><td data-label=\"Assumption\">Hazard and exposure are independent within an event<\/td><td data-label=\"Why it is made\">Keeps the loss integral tractable<\/td><td data-label=\"How it is tested\">Compare with a copula fit on joint records<\/td><\/tr><tr><td data-label=\"Assumption\">Couplings are linear near the operating point<\/td><td data-label=\"Why it is made\">Lets shocks propagate through A<\/td><td data-label=\"How it is tested\">Perturb beyond the linear range and compare<\/td><\/tr><tr><td data-label=\"Assumption\">Vulnerability curves transfer between places<\/td><td data-label=\"Why it is made\">Local curves are rarely available<\/td><td data-label=\"How it is tested\">Validate against observed losses where they exist<\/td><\/tr><\/tbody><\/table><\/div><dl class=\"nx-au-kv\"><div><dt>Inputs<\/dt><dd>versioned datasets with provenance<\/dd><\/div><div><dt>Outputs<\/dt><dd>distributions, never single numbers<\/dd><\/div><div><dt>Code<\/dt><dd>released with the model card and tests<\/dd><\/div><div><dt>Reproduction<\/dt><dd>same data, code, environment and seed<\/dd><\/div><\/dl><figcaption class=\"nx-caption\"><span>Standard risk and state-space formulations; the assumptions listed are typical of the method, stated for review, not results.<\/span><\/figcaption><\/figure><\/div><div class=\"nx-chips nx-au-fw\" aria-label=\"Frameworks cited\"><span class=\"nx-chip\">FAIR guiding principles \u00b7 GO FAIR and FORCE11<\/span><span class=\"nx-chip\">W3C PROV \u00b7 World Wide Web Consortium<\/span><span class=\"nx-chip\">Guidance on consistent treatment of uncertainties \u00b7 IPCC<\/span><span class=\"nx-chip\">CEOS Analysis Ready Data \u00b7 Committee on Earth Observation Satellites<\/span><\/div><\/div><\/div><\/section><\/div><\/div><\/section><section class=\"nx-section\" data-band=\"auto\" data-band-id=\"nx-band-2\" data-tone=\"alt\" data-hue=\"infrastructure\" aria-label=\"Validation, Connections, Stewardship, Questions\"><div class=\"nx-container\"><div class=\"nx-hm-band nx-band\" data-nx-tabs=\"true\" data-active=\"p1\"><div class=\"nx-tablist nx-hm-tabs\" role=\"tablist\" aria-label=\"Validation, Connections, Stewardship, Questions\"><button class=\"nx-tab\" type=\"button\" role=\"tab\" id=\"nx-band-2-tab-1\" aria-controls=\"nx-sec-checks\" aria-selected=\"true\" data-key=\"p1\">Validation<\/button><button class=\"nx-tab\" type=\"button\" role=\"tab\" id=\"nx-band-2-tab-2\" aria-controls=\"nx-sec-connections\" aria-selected=\"false\" data-key=\"p2\">Connections<\/button><button class=\"nx-tab\" type=\"button\" role=\"tab\" id=\"nx-band-2-tab-3\" aria-controls=\"nx-sec-stewardship\" aria-selected=\"false\" data-key=\"p3\">Stewardship<\/button><button class=\"nx-tab\" type=\"button\" role=\"tab\" id=\"nx-band-2-tab-4\" aria-controls=\"nx-s-9-sec\" aria-selected=\"false\" data-key=\"p4\">Questions<\/button><\/div><section id=\"nx-sec-checks\" class=\"nx-section nx-band-panel\" data-hue=\"infrastructure\" aria-labelledby=\"nx-s-6\" role=\"tabpanel\" data-band-tab=\"nx-band-2-tab-1\"><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\">Validation<\/span><\/div><h2 id=\"nx-s-6\" class=\"nx-section-title\">How results are checked.<\/h2><p class=\"nx-lede\">Models are evaluated on held-out data that respects spatial, temporal and group structure. Robustness, calibration and fairness are tested alongside accuracy.<\/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\">Check<\/th><th scope=\"col\">How it is done<\/th><\/tr><\/thead><tbody><tr><td data-label=\"Check\">Held-out evaluation<\/td><td data-label=\"How it is done\">Models are evaluated on data not used for training or tuning<\/td><\/tr><tr><td data-label=\"Check\">Dataset documentation<\/td><td data-label=\"How it is done\">Training data carry datasheets on origin, consent and limits<\/td><\/tr><tr><td data-label=\"Check\">Drift monitoring<\/td><td data-label=\"How it is done\">Deployed models are monitored for input and performance drift<\/td><\/tr><\/tbody><\/table><\/div><p class=\"nx-note nx-mt-3\">Performance degrades when deployment conditions differ from training data. Models can encode biases and fail without clear warning on rare events.<\/p><\/div><\/div><\/section><section id=\"nx-sec-connections\" class=\"nx-section nx-band-panel\" data-hue=\"project\" data-variant=\"split\" aria-labelledby=\"nx-s-7\" role=\"tabpanel\" data-band-tab=\"nx-band-2-tab-2\"><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\">Connections<\/span><\/div><h2 id=\"nx-s-7\" class=\"nx-section-title\">Topics that rely on this method.<\/h2><\/div><div class=\"nx-section-body\"><ul class=\"nx-list\" data-nav=\"true\"><li><a class=\"nx-link\" href=\"https:\/\/therisk.global\/ai-autonomous-systems\/topics\/model-evaluation\/\"><span>Model evaluation<\/span><svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><\/li><li><a class=\"nx-link\" href=\"https:\/\/therisk.global\/ai-autonomous-systems\/topics\/agentic-systems\/\"><span>Agentic systems<\/span><svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><\/li><li><a class=\"nx-link\" href=\"https:\/\/therisk.global\/ai-autonomous-systems\/topics\/ai-cyber-security\/\"><span>AI cyber security<\/span><svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><\/li><li><a class=\"nx-link\" href=\"https:\/\/therisk.global\/ai-autonomous-systems\/topics\/ai-in-science\/\"><span>AI in science<\/span><svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><\/li><li><a class=\"nx-link\" href=\"https:\/\/therisk.global\/ai-autonomous-systems\/topics\/ai-and-public-services\/\"><span>AI and public services<\/span><svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><\/li><li><a class=\"nx-link\" href=\"https:\/\/therisk.global\/ai-autonomous-systems\/topics\/autonomous-physical-systems\/\"><span>Autonomous physical systems<\/span><svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><\/li><li><a class=\"nx-link\" href=\"https:\/\/therisk.global\/ai-autonomous-systems\/topics\/compute-and-energy\/\"><span>Compute and energy<\/span><svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><\/li><li><a class=\"nx-link\" href=\"https:\/\/therisk.global\/ai-autonomous-systems\/topics\/labour-and-social-effects\/\"><span>Labour and social effects<\/span><svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><\/li><\/ul><\/div><\/div><\/section><section id=\"nx-sec-stewardship\" class=\"nx-section nx-band-panel\" data-hue=\"infrastructure\" aria-labelledby=\"nx-s-8\" role=\"tabpanel\" data-band-tab=\"nx-band-2-tab-3\"><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\">Stewardship<\/span><\/div><h2 id=\"nx-s-8\" class=\"nx-section-title\">Who maintains the method.<\/h2><p class=\"nx-lede\">Artificial Intelligence and Autonomous <a class=\"nx-link nx-term\" href=\"https:\/\/docs.therisk.global\/organization\/systems\" data-tc=\"Systems\" data-tk=\"Nexus Standards\" data-td=\"The twenty Nexus systems, NXS-CMD to NXS-LAB\">Systems<\/a> stewards artificial intelligence and machine learning for the Consortium and reviews how it is applied across platforms.<\/p><\/div><div class=\"nx-section-body\"><p class=\"nx-p\"><a class=\"nx-link\" href=\"https:\/\/therisk.global\/stem\/technology\/ai-ml\/\">Canonical page: Artificial intelligence and machine learning<\/a><\/p><\/div><\/div><\/section><section id=\"nx-s-9-sec\" class=\"nx-section nx-band-panel\" data-hue=\"project\" aria-labelledby=\"nx-s-9\" role=\"tabpanel\" data-band-tab=\"nx-band-2-tab-4\"><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\">Questions<\/span><\/div><h2 id=\"nx-s-9\" class=\"nx-section-title\">Questions and answers.<\/h2><\/div><div class=\"nx-section-body\"><div class=\"nx-faq\"><details><summary>What are the limits of artificial intelligence and machine learning in this field?<\/summary><div class=\"nx-answer\">Performance degrades when deployment conditions differ from training data. Models can encode biases and fail without clear warning on rare events.<\/div><\/details><details><summary>How are results checked?<\/summary><div class=\"nx-answer\">Models are evaluated on held-out data that respects spatial, temporal and group structure. Robustness, calibration and fairness are tested alongside accuracy.<\/div><\/details><details><summary>Are the values on this page real?<\/summary><div class=\"nx-answer\">The data names, roles and checks are real descriptions of practice. The canvas shows a design of the method&#8217;s use, not a run; no result is reported here.<\/div><\/details><details><summary>Where else is this method applied?<\/summary><div class=\"nx-answer\">The <a class=\"nx-link\" href=\"https:\/\/therisk.global\/stem\/technology\/ai-ml\/\">canonical method page<\/a> lists every platform that applies it.<\/div><\/details><\/div><\/div><\/div><\/section><\/div><\/div><\/section><section id=\"nx-sec-nr-gateway\" class=\"nx-section\" data-hue=\"infrastructure\" data-tone=\"alt\" aria-labelledby=\"nx-nr-gateway-h\"><div class=\"nx-container\"><div class=\"nx-section-head\"><div class=\"nx-eyebrow-row\"><span class=\"nx-eyebrow\">Reports<\/span><span class=\"nx-rule\" aria-hidden=\"true\"><\/span><span class=\"nx-eyebrow\">Contribute<\/span><\/div><h2 id=\"nx-nr-gateway-h\" class=\"nx-section-title\">Contribute evidence from this page.<\/h2><p class=\"nx-lede\"><a class=\"nx-link\" href=\"https:\/\/docs.therisk.global\/organization\/programme\/reports\">Reports<\/a> is the Consortium&#8217;s quarterly publication. What you know about this subject can become evidence, a case or a correction in the next edition, credited to you and published under CC BY 4.0 where it is your own work.<\/p><div class=\"nx-chips\"><span class=\"nx-chip\">GCRI<\/span><span class=\"nx-chip\">Artificial Intelligence and Autonomous Systems<\/span><span class=\"nx-chip\">Exponential Technology<\/span><span class=\"nx-chip\">Labs Evidence<\/span><\/div><\/div><div class=\"nx-section-body\"><ul class=\"nx-dest\"><li><a class=\"nx-dest-card\" href=\"https:\/\/therisk.global\/ai-autonomous-systems\/write-to-us\/contact\/\"><span class=\"nx-dest-top\"><span class=\"nx-dest-rel\">Email<\/span><span class=\"nx-dest-host\">Gateway GW-GCRI<\/span><\/span><span class=\"nx-dest-title\">Send it from this page<\/span><span class=\"nx-dest-why\">Evidence, a correction, a case, a method, a review, a translation or profile input. It arrives tagged with this page, so the editor who owns it receives it.<\/span><span class=\"nx-dest-go\" aria-hidden=\"true\"><svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/span><\/a><\/li><li><a class=\"nx-dest-card\" href=\"https:\/\/therisk.global\/nexus-reports\/open-science\/#nx-sec-zenodo\"><span class=\"nx-dest-top\"><span class=\"nx-dest-rel\">Zenodo<\/span><span class=\"nx-dest-host\">Open deposit<\/span><\/span><span class=\"nx-dest-title\">Deposit it and submit to the community<\/span><span class=\"nx-dest-why\">Publish your own record on Zenodo with your ORCID iD, then submit it to the Reports community there. Curators review it; the DOI stays yours.<\/span><span class=\"nx-dest-go\" aria-hidden=\"true\"><svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/span><\/a><\/li><li><a class=\"nx-dest-card\" href=\"https:\/\/therisk.global\/nexus-reports\/contribute\/\"><span class=\"nx-dest-top\"><span class=\"nx-dest-rel\">Reports<\/span><span class=\"nx-dest-host\">GCRI<\/span><\/span><span class=\"nx-dest-title\">See every way to contribute<\/span><span class=\"nx-dest-why\">The contribution types, how each is reviewed, what credit you receive, the licence and what is never accepted.<\/span><span class=\"nx-dest-go\" aria-hidden=\"true\"><svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/span><\/a><\/li><\/ul><div class=\"nx-live-slot\" data-nxf-slot=\"contribute\" data-nxf-kind=\"workflow\" data-nxf-resource=\"nexus:\/\/gcri\/ai-autonomous-systems\/stem\/ai-ml#contribute\" data-nxf-title=\"Contribute evidence\" data-nxf-height=\"480\" hidden><\/div><p class=\"nx-note nx-mt-3\">Never send personal data, sensitive infrastructure detail or anything you have no right to share. A contribution is reviewed before use; sending it confers no authorship or editorial rights, and sponsorship never buys a finding.<\/p><\/div><\/div><\/section><section class=\"nx-section nx-deeper-band\" aria-labelledby=\"nx-deeper-therisk-global-ai-autonomous-systems-stem-ai-ml-h\"><div class=\"nx-container\"><nav class=\"nx-deeper\" aria-labelledby=\"nx-deeper-therisk-global-ai-autonomous-systems-stem-ai-ml-h\" data-nx-deeper=\"3\" data-nx-resources=\"1 1 1\"><h2 class=\"nx-deeper-h\" id=\"nx-deeper-therisk-global-ai-autonomous-systems-stem-ai-ml-h\">Go deeper<\/h2><ul><li><a href=\"https:\/\/docs.therisk.global\/organization\/systems\/nxs-dss\" data-ext><span class=\"nx-deeper-k\">Documentation:<\/span> Nxs-dss<span class=\"nx-deeper-x\" aria-hidden=\"true\">\u2197<\/span><span class=\"nx-deeper-sr\"> (opens another site)<\/span><\/a><\/li><li><a href=\"https:\/\/risksacademy.com\/technology-digital\/artificial-intelligence\/\" data-ext><span class=\"nx-deeper-k\">Academy:<\/span> Artificial intelligence and machine learning<span class=\"nx-deeper-x\" aria-hidden=\"true\">\u2197<\/span><span class=\"nx-deeper-sr\"> (opens another site)<\/span><\/a><\/li><li><a href=\"https:\/\/nexustandards.com\/digital-ai\/cyber-and-digital-risk-standards\/\" data-ext><span class=\"nx-deeper-k\">Standards:<\/span> Cyber and digital risk standards<span class=\"nx-deeper-x\" aria-hidden=\"true\">\u2197<\/span><span class=\"nx-deeper-sr\"> (opens another site)<\/span><\/a><\/li><\/ul><\/nav><\/div><\/section><section class=\"nx-closing\" aria-labelledby=\"nx-close-10\"><div class=\"nx-container\"><span class=\"nx-eyebrow\">Artificial Intelligence and Autonomous Systems<\/span><h2 id=\"nx-close-10\" class=\"nx-section-title\">Bring a method, a dataset or a question.<\/h2><p class=\"nx-lede\">Information and enquiries are open today. 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=\"https:\/\/therisk.global\/ai-autonomous-systems\/write-to-us\/research-collaboration\/\">Propose a method <svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\"><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><section class=\"nx-section\" data-tone=\"alt\" aria-labelledby=\"nx-legal-11\"><div class=\"nx-container\"><span class=\"nx-eyebrow\">Legal<\/span><h2 id=\"nx-legal-11\" class=\"nx-h3 nx-mt-2\">Legal documents.<\/h2><nav class=\"nx-related nx-mt-3\" aria-label=\"Legal documents\"><a href=\"https:\/\/therisk.global\/legal\/terms\/\">Platform Terms of Use <svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><a href=\"https:\/\/therisk.global\/legal\/privacy\/\">Platform Privacy Notice <svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><a href=\"https:\/\/therisk.global\/legal\/cookies\/\">Platform Cookie and Storage Notice <svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><a href=\"https:\/\/therisk.global\/legal\/legal-notice\/\">Legal Notice <svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><a href=\"https:\/\/therisk.global\/legal\/acceptable-use\/\">Acceptable Use Policy <svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><a href=\"https:\/\/therisk.global\/legal\/participation-terms\/\">Participation and Events Terms <svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><a href=\"https:\/\/therisk.global\/legal\/contribution-terms\/\">Contribution Terms <svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><a href=\"https:\/\/therisk.global\/legal\/research-and-data-terms\/\">Research and Data Terms <svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><a href=\"https:\/\/therisk.global\/legal\/ai-use\/\">AI-use Statement <svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><a href=\"https:\/\/therisk.global\/legal\/intellectual-property\/\">Intellectual Property and Licensing <svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><a href=\"https:\/\/therisk.global\/legal\/trademark-policy\/\">Trademark and Name Use Policy <svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><a href=\"https:\/\/therisk.global\/legal\/disclaimer\/\">Disclaimers and Data Notices <svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><a href=\"https:\/\/therisk.global\/legal\/corrections\/\">Corrections and Right of Reply Policy <svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><a href=\"https:\/\/therisk.global\/legal\/complaints-and-integrity\/\">Complaints and integrity reporting <svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><a href=\"https:\/\/therisk.global\/legal\/security\/\">Security and responsible disclosure <svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><a href=\"https:\/\/therisk.global\/legal\/code-of-conduct\/\">Code of Conduct <svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><a href=\"https:\/\/therisk.global\/trust\/accessibility\/\">Accessibility Statement <svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><a href=\"https:\/\/therisk.global\/legal\/third-party-notices\/\">Third-party notices and attributions <svg class=\"nx-arrow\" viewBox=\"0 0 16 16\" aria-hidden=\"true\"><path d=\"M3 8h10M9 4l4 4-4 4\"\/><\/svg><\/a><\/nav><p class=\"nx-note nx-mt-3\">Maps, data and figures on this site are shown for context and general information. They are not advice, forecasts or warnings; the limits of the information are set out in the Disclaimers and Data Notices.<\/p><\/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\":\"TechArticle\",\"@id\":\"https:\/\/therisk.global\/ai-autonomous-systems\/stem\/ai-ml\/#article\",\"headline\":\"Machine learning behaviour\",\"description\":\"Machine learning generalisation: distribution shift and drift, fine-tuning for a specific use and interpretability methods with their limits stated.\",\"inLanguage\":\"en\",\"about\":{\"@type\":\"Thing\",\"name\":\"machine learning generalisation\"},\"author\":{\"@id\":\"https:\/\/therisk.global\/#organization\",\"@type\":\"Organization\",\"name\":\"The Global Centre for Risk and Innovation\",\"url\":\"https:\/\/therisk.global\/\",\"alternateName\":\"GCRI\"},\"publisher\":{\"@id\":\"https:\/\/therisk.global\/#organization\"},\"mainEntityOfPage\":{\"@id\":\"https:\/\/therisk.global\/ai-autonomous-systems\/stem\/ai-ml\/#webpage\"},\"isAccessibleForFree\":true,\"mentions\":[{\"@type\":\"Thing\",\"name\":\"Training and validation datasets with documentation\"},{\"@type\":\"Thing\",\"name\":\"Deployment input logs\"},{\"@type\":\"Thing\",\"name\":\"Benchmark suites\"}]}]}<\/script>\n\n<!-- nx-rev:08e48d174d710380 -->\n<!-- nx-shared:60e0eacbf34a -->\n<!-- nx-path:\/stem\/ai-ml\/ -->","protected":false},"excerpt":{"rendered":"<p>40){clearInterval(re)}},400);n.NXLive={allowed:d,blocked:f,grant:B,register:V,hosts:E.length}})();GCRI \u00b7 Artificial Intelligence and Autonomous Systems<\/p>\n","protected":false},"author":2,"featured_media":0,"parent":25,"menu_order":325,"comment_status":"closed","ping_status":"closed","template":"","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":"","_nx_seo_description":"Machine learning generalisation: distribution shift and drift, fine-tuning for a specific use and interpretability methods with their limits stated.","_nx_seo_title":"Machine learning generalisation | GCRI","_nx_seo_image":"","_nx_seo_jsonld":"{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"Organization\",\"@id\":\"https:\/\/therisk.global\/#organization\",\"name\":\"The Global Centre for Risk and Innovation\",\"alternateName\":\"GCRI\",\"url\":\"https:\/\/therisk.global\/\",\"description\":\"The Global Centre for Risk and Innovation (GCRI) works on risk science, evidence, methods and technical capability. GCRI is a research, development and policy institution. It is not a regulator, and it does not certify, license or approve; its findings carry their methods and uncertainty.\",\"address\":{\"@type\":\"PostalAddress\",\"addressLocality\":\"Toronto\",\"addressRegion\":\"ON\",\"postalCode\":\"M4R 1V2\",\"addressCountry\":\"CA\"},\"knowsAbout\":[\"Systemic risk\",\"Disaster risk reduction\",\"Risk science\",\"Earth observation\",\"Water security\",\"Food security\",\"Energy systems\",\"Global health\",\"Biodiversity\",\"Critical infrastructure\"],\"contactPoint\":[{\"@type\":\"ContactPoint\",\"contactType\":\"Consortium Secretariat\",\"url\":\"https:\/\/therisk.global\/write-to-us\/contact\/\",\"availableLanguage\":[\"en\"]}],\"subOrganization\":[{\"@type\":\"Organization\",\"name\":\"GCRI Canada\",\"url\":\"https:\/\/therisk.global\/about\/gcri-canada\/\",\"description\":\"Lead organisation of GCRI: Consortium Secretariat and global contracting party in support of the national and regional consortiums.\",\"address\":{\"@type\":\"PostalAddress\",\"addressLocality\":\"Toronto\",\"addressRegion\":\"ON\",\"postalCode\":\"M4R 1V2\",\"addressCountry\":\"CA\"},\"parentOrganization\":{\"@id\":\"https:\/\/therisk.global\/#organization\"}},{\"@type\":\"Organization\",\"name\":\"GCRI US\",\"url\":\"https:\/\/therisk.global\/about\/gcri-us\/\",\"description\":\"The United States institution of GCRI, a separate organisation with its own legal identity.\",\"address\":{\"@type\":\"PostalAddress\",\"streetAddress\":\"4500 Forbes Blvd., Suite 200\",\"addressLocality\":\"Lanham\",\"addressRegion\":\"MD\",\"postalCode\":\"20706\",\"addressCountry\":\"US\"},\"parentOrganization\":{\"@id\":\"https:\/\/therisk.global\/#organization\"}}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/therisk.global\/ai-autonomous-systems\/#website\",\"url\":\"https:\/\/therisk.global\/ai-autonomous-systems\/\",\"name\":\"Artificial Intelligence and Autonomous Systems\",\"publisher\":{\"@id\":\"https:\/\/therisk.global\/#organization\"},\"inLanguage\":\"en\"},{\"@type\":\"WebPage\",\"@id\":\"https:\/\/therisk.global\/ai-autonomous-systems\/stem\/ai-ml\/#webpage\",\"url\":\"https:\/\/therisk.global\/ai-autonomous-systems\/stem\/ai-ml\/\",\"name\":\"Machine learning behaviour\",\"headline\":\"Machine learning generalisation | GCRI\",\"description\":\"Machine learning generalisation: distribution shift and drift, fine-tuning for a specific use and interpretability methods with their limits stated.\",\"inLanguage\":\"en\",\"isPartOf\":{\"@id\":\"https:\/\/therisk.global\/ai-autonomous-systems\/#website\"},\"publisher\":{\"@id\":\"https:\/\/therisk.global\/#organization\"},\"keywords\":\"machine learning generalisation, distribution shift, fine-tuning, interpretability, model drift, training and validation datasets with documentation\",\"about\":{\"@type\":\"Thing\",\"name\":\"machine learning generalisation\"},\"breadcrumb\":{\"@id\":\"https:\/\/therisk.global\/ai-autonomous-systems\/stem\/ai-ml\/#breadcrumb\"}},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/therisk.global\/ai-autonomous-systems\/stem\/ai-ml\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"GCRI\",\"item\":\"https:\/\/therisk.global\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Artificial Intelligence and Autonomous Systems\",\"item\":\"https:\/\/therisk.global\/ai-autonomous-systems\/\"},{\"@type\":\"ListItem\",\"position\":3,\"name\":\"STEM foundation of Artificial Intelligence and Autonomous Systems\",\"item\":\"https:\/\/therisk.global\/ai-autonomous-systems\/stem\/\"},{\"@type\":\"ListItem\",\"position\":4,\"name\":\"Machine learning behaviour\",\"item\":\"https:\/\/therisk.global\/ai-autonomous-systems\/stem\/ai-ml\/\"}]},{\"@type\":\"TechArticle\",\"@id\":\"https:\/\/therisk.global\/ai-autonomous-systems\/stem\/ai-ml\/#article\",\"headline\":\"Machine learning behaviour\",\"description\":\"Machine learning generalisation: distribution shift and drift, fine-tuning for a specific use and interpretability methods with their limits stated.\",\"inLanguage\":\"en\",\"about\":{\"@type\":\"Thing\",\"name\":\"machine learning generalisation\"},\"author\":{\"@id\":\"https:\/\/therisk.global\/#organization\"},\"publisher\":{\"@id\":\"https:\/\/therisk.global\/#organization\"},\"mainEntityOfPage\":{\"@id\":\"https:\/\/therisk.global\/ai-autonomous-systems\/stem\/ai-ml\/#webpage\"},\"isAccessibleForFree\":true,\"mentions\":[{\"@type\":\"Thing\",\"name\":\"Training and validation datasets with documentation\"},{\"@type\":\"Thing\",\"name\":\"Deployment input logs\"},{\"@type\":\"Thing\",\"name\":\"Benchmark suites\"}]}]}","_nx_i18n_locale":"","_nx_i18n_group":"","nexus_page_kind":"","footnotes":"","_nx_seo_image_alt":"","_nx_seo_canonical":"https:\/\/therisk.global\/ai-autonomous-systems\/stem\/ai-ml\/","_nx_seo_robots":"index,follow","_nx_seo_schema_type":"TechArticle"},"class_list":["post-39","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/therisk.global\/ai-autonomous-systems\/wp-json\/wp\/v2\/pages\/39","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\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/therisk.global\/ai-autonomous-systems\/wp-json\/wp\/v2\/comments?post=39"}],"version-history":[{"count":1,"href":"https:\/\/therisk.global\/ai-autonomous-systems\/wp-json\/wp\/v2\/pages\/39\/revisions"}],"predecessor-version":[{"id":94,"href":"https:\/\/therisk.global\/ai-autonomous-systems\/wp-json\/wp\/v2\/pages\/39\/revisions\/94"}],"up":[{"embeddable":true,"href":"https:\/\/therisk.global\/ai-autonomous-systems\/wp-json\/wp\/v2\/pages\/25"}],"wp:attachment":[{"href":"https:\/\/therisk.global\/ai-autonomous-systems\/wp-json\/wp\/v2\/media?parent=39"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}