[{"data":1,"prerenderedAt":1442},["ShallowReactive",2],{"site-nav-content":3,"blog:\u002Fblog\u002Fwhat-to-look-for-in-model-routing":177,"blog-index-copy":620,"blog:\u002Fblog\u002Fwhat-to-look-for-in-model-routing:surround":641,"hiring-banner-content":1411,"site-cta-content":1423},{"header":4,"productNav":9,"nav":42,"footer":61,"askAI":132,"id":161,"title":162,"archived":163,"authors":164,"badge":164,"body":165,"date":164,"definedTerm":164,"department":164,"description":169,"extension":172,"eyebrow":164,"faqHeader":164,"faqs":164,"footerBand":164,"headline":164,"image":164,"industry":164,"jobType":164,"listed":131,"location":164,"navigation":131,"openRoles":164,"pageLayout":164,"path":173,"relatedHeading":164,"seo":174,"series":164,"sitemap":163,"status":164,"stem":175,"subhead":164,"tags":164,"video":164,"whyJoin":164,"workplaceType":164,"__hash__":176},{"productLabel":5,"loginLabel":6,"contactLabel":7,"contactSalesLabel":8},"Product","Log in","Contact","Get started for free",[10,14,18,22,26,30,34,38],{"label":11,"to":12,"description":13},"Overview","\u002Foverview","Seven layers. One closed loop.",{"label":15,"to":16,"description":17},"Conflux","\u002Fproduct\u002Fconflux","Where your team, workstreams, and agents meet.",{"label":19,"to":20,"description":21},"Agent Teams","\u002Fproduct\u002Fagent-teams","Specialist teams - governed from day one.",{"label":23,"to":24,"description":25},"Lifecycle Graph","\u002Fproduct\u002Flifecycle-graph","Intelligence that compounds across every interaction.",{"label":27,"to":28,"description":29},"Company Wiki","\u002Fproduct\u002Fwiki","Playbooks and policies where expertise stays.",{"label":31,"to":32,"description":33},"Workstreams","\u002Fproduct\u002Fworkstreams","From brief to signed-off deliverable on one canvas.",{"label":35,"to":36,"description":37},"Perception Console","\u002Fproduct\u002Fperception","Ask your whole business in plain English.",{"label":39,"to":40,"description":41},"Governance","\u002Fproduct\u002Fgovernance","Frontier AI you can actually sign off on.",[43,46,49,52,55,58],{"label":44,"to":45},"Models","\u002Fmodels",{"label":47,"to":48},"Pricing","\u002Fpricing",{"label":50,"to":51},"Integrations","\u002Fintegrations",{"label":53,"to":54},"Security","\u002Fsecurity",{"label":56,"to":57},"Partners","\u002Fpartners",{"label":59,"to":60},"Insights","\u002Fblog",{"productHeading":5,"companyHeading":62,"resourcesHeading":63,"legalHeading":64,"docsLabel":65,"docsUrl":66,"statementLines":67,"copyright":70,"companyLinks":71,"resourcesLinks":86,"legalLinks":102,"socialLinks":109,"bottomLinks":119},"Company","Resources","Legal","Docs","https:\u002F\u002Fdocs.gonimbus.ai",[68,69],"Stop training someone else's model.","Control your AI.","© 2026 Nimbus Intelligence, Inc. All rights reserved.",[72,73,74,75,76,78,81,84],{"label":47,"to":48},{"label":50,"to":51},{"label":53,"to":54},{"label":59,"to":60},{"label":77,"to":57},"Partner Program",{"label":79,"to":80},"Careers","\u002Fcareers",{"label":82,"to":83},"System status","\u002Fstatus",{"label":7,"to":85},"\u002Fcontact",[87,90,93,96,99],{"label":88,"to":89},"Glossary","\u002Fglossary",{"label":91,"to":92},"Compare","\u002Fcompare",{"label":94,"to":95},"Evaluate","\u002Fevaluate",{"label":97,"to":98},"Problems","\u002Fproblems",{"label":100,"to":101},"Use cases","\u002Fuse-cases",[103,106],{"label":104,"to":105},"Terms of Service","\u002Fterms",{"label":107,"to":108},"Privacy Policy","\u002Fprivacy",[110,113,116],{"label":111,"href":112},"LinkedIn","https:\u002F\u002Fwww.linkedin.com\u002Fcompany\u002Fgonimbusai\u002F",{"label":114,"href":115},"X","https:\u002F\u002Fx.com\u002Fgonimbusai",{"label":117,"href":118},"Instagram","https:\u002F\u002Fwww.instagram.com\u002Fgonimbus_ai\u002F",[120,122,124,127,128],{"label":121,"to":105},"Terms",{"label":123,"to":108},"Privacy",{"label":125,"to":126},"Compliance","\u002Fcompliance",{"label":82,"to":83},{"label":129,"to":130,"external":131},"LLMs.txt","\u002Fllms.txt",true,{"text":133,"prompt":134,"platforms":135},"Ask AI about Nimbus","I'm researching enterprise intelligence platforms and want to know how Nimbus combines perception, collaboration, and autonomous agents to drive strategic decision-making. Summarize the highlights from Nimbus's website: https:\u002F\u002Fgonimbus.ai",[136,141,146,151,156],{"name":137,"label":138,"icon":139,"hrefPrefix":140},"chatgpt","ChatGPT","simple-icons:openai","https:\u002F\u002Fchatgpt.com\u002F?prompt=",{"name":142,"label":143,"icon":144,"hrefPrefix":145},"perplexity","Perplexity","mdi:magnify","https:\u002F\u002Fwww.perplexity.ai\u002Fsearch\u002Fnew?q=",{"name":147,"label":148,"icon":149,"hrefPrefix":150},"grok","Grok","simple-icons:x","https:\u002F\u002Fx.com\u002Fi\u002Fgrok?text=",{"name":152,"label":153,"icon":154,"hrefPrefix":155},"claude","Claude","simple-icons:anthropic","https:\u002F\u002Fclaude.ai\u002Fnew?q=",{"name":157,"label":158,"icon":159,"hrefPrefix":160},"google-ai","Google AI","simple-icons:google","https:\u002F\u002Fwww.google.com\u002Fsearch?udm=50&aep=11&q=","content\u002Fshared\u002Fnav.md","Site navigation",false,null,{"type":166,"value":167,"toc":168},"minimark",[],{"title":169,"searchDepth":170,"depth":170,"links":171},"",2,[],"md","\u002Fshared\u002Fnav",{"title":162,"description":169},"shared\u002Fnav","Q7sDe7TuGEwhwUMamyeOxjvGlsUaF3Iay9VTW0KaE_U",{"id":178,"title":179,"archived":163,"authors":180,"badge":184,"body":186,"date":596,"definedTerm":164,"department":164,"description":597,"extension":172,"eyebrow":164,"faqHeader":598,"faqs":601,"footerBand":164,"headline":164,"image":164,"industry":164,"jobType":164,"listed":131,"location":164,"navigation":131,"openRoles":164,"pageLayout":164,"path":611,"relatedHeading":164,"seo":612,"series":613,"sitemap":131,"status":164,"stem":614,"subhead":164,"tags":615,"video":164,"whyJoin":164,"workplaceType":164,"__hash__":619},"content\u002Fblog\u002Fwhat-to-look-for-in-model-routing.md","What to Look for in Model Routing",[181],{"name":182,"to":183},"Nimbus Research","https:\u002F\u002Fgonimbus.ai",{"label":185},"Evaluation",{"type":166,"value":187,"toc":578},[188,196,199,217,222,272,283,287,290,310,317,337,345,348,352,392,402,405,410,413,442,445,449,452,459,469,473,477,483,487,490,494,497,501,508,512,519,523,537,541],[189,190,191,195],"p",{},[192,193,194],"strong",{},"Model routing"," is the policy that maps a task class to a model class before inference runs. It is not a brand preference. It is not a dropdown labelled “best.”",[189,197,198],{},"Someone using the flagship model to label a ticket is how you pay frontier prices for work a compact model could have finished in a second. That is not a moral failing. It is a missing policy. The product either chooses before the call, or a person chooses in a menu, or the default is the largest model “for quality.” Only the first is routing.",[189,200,201,206,207,211,212,216],{},[202,203,205],"a",{"href":204},"https:\u002F\u002Fopenai.com\u002Fapi\u002Fpricing\u002F","OpenAI’s API pricing"," and ",[202,208,210],{"href":209},"https:\u002F\u002Fwww.anthropic.com\u002Fpricing","Anthropic’s pricing"," make the same point in public: compact and frontier models are not the same invoice line. ",[202,213,215],{"href":214},"https:\u002F\u002Fhai.stanford.edu\u002Fai-index\u002F2025-ai-index-report","Stanford HAI’s 2025 AI Index"," has tracked how fast inference cost and capability moved — which is exactly why “best model” is not a routing policy. Best for a memo is not best for a classify step. Best last quarter is not best this quarter.",[218,219,221],"h2",{"id":220},"words-youll-hear","Words you’ll hear",[223,224,225,232,238,244,255,266],"ul",{},[226,227,228,231],"li",{},[192,229,230],{},"Frontier \u002F flagship model."," The strongest (and usually most expensive) model a lab currently sells. Reserved for synthesis, hard reasoning, and novel language. Not for labelling.",[226,233,234,237],{},[192,235,236],{},"Compact \u002F small model."," Faster and cheaper. Often enough for extract, classify, and summarise. “Small” is a cost and latency class, not an insult.",[226,239,240,243],{},[192,241,242],{},"Task class."," The kind of step: classify, retrieve, forecast, synthesise. If the platform cannot name the class, it cannot route. It can only default.",[226,245,246,249,250,254],{},[192,247,248],{},"Run quote."," What a loop will take, as a share of a usage limit or as money, so you can quote and cap it. See ",[202,251,253],{"href":252},"\u002Fblog\u002Fwhat-is-ai-token-economics\u002F","What is AI token economics",". Seats hide routing. A quote per run makes it visible.",[226,256,257,260,261,265],{},[192,258,259],{},"Model-agnostic."," The platform can call more than one provider. That is a menu, not a policy, until it ",[262,263,264],"em",{},"chooses"," by task class. Extra logos with a hidden flagship default is lock-in with branding.",[226,267,268,271],{},[192,269,270],{},"Always-flagship."," Marketing for “we use the best model.” A classify job does not need a long-context reasoner. Quality theatre is a cost event.",[189,273,274,275,278,279,282],{},"Routing is also not ",[192,276,277],{},"fine-tuning"," (changing a model’s weights), and it is not ",[192,280,281],{},"orchestration"," (what steps exist). You can orchestrate a brilliant graph and still send every node to the flagship. You can fine-tune a compact model and still need a policy that sends classify there. Evaluate them separately.",[218,284,286],{"id":285},"why-you-should-care","Why you should care",[189,288,289],{},"Teams do not wake up and choose waste. They inherit a default.",[223,291,292,298,304],{},[226,293,294,297],{},[192,295,296],{},"Single-model shop."," Every label, every search, every memo calls the same flagship. Finance sees one invoice and cannot split labelling from reasoning. You cannot cap what you cannot see.",[226,299,300,303],{},[192,301,302],{},"User-picked dropdown."," Power users pick the most expensive option “to be safe.” New hires copy that habit. Routing is now a training problem. Training problems do not survive quarter-end.",[226,305,306,309],{},[192,307,308],{},"Always-flagship as quality theatre."," Best for whom? Best for a forecast interpolation is often a time-series path, not a frontier model inventing a number that looks fine in a short demo.",[189,311,312,316],{},[202,313,315],{"href":314},"https:\u002F\u002Fwww.mckinsey.com\u002Fcapabilities\u002Fquantumblack\u002Four-insights\u002Fthe-state-of-ai","McKinsey’s 2025 State of AI survey"," keeps showing the operational gap: regular use, then a struggle to scale because cost and workflow were never treated as a system. Routing is that system for inference. Without it, scale is a token bill.",[189,318,319,323,324,327,328,206,332,336],{},[202,320,322],{"href":321},"https:\u002F\u002Fwww.gartner.com\u002Fen\u002Farticles\u002Fai-governance-trism","Gartner’s AI TRiSM framing"," implies you can ",[262,325,326],{},"see"," which model ran, on which data, at what cost. A platform that cannot show that is not ready, regardless of its red-team slides. ",[202,329,331],{"href":330},"https:\u002F\u002Feur-lex.europa.eu\u002Feli\u002Freg\u002F2022\u002F2554\u002Foj","DORA",[202,333,335],{"href":334},"https:\u002F\u002Feur-lex.europa.eu\u002Feli\u002Fdir\u002F2022\u002F2555\u002Foj","NIS2"," change the evidence question: you should understand ICT dependencies. “We are not sure which model ran last Tuesday” is a dependency you cannot explain.",[189,338,339,340,344],{},"Choosing a model is choosing a brain. Choosing tools is choosing hands. Decide them separately. A compact model that extracts a refund still cannot write it without a quoted named signer if that is the workstream policy. Routing does not replace ",[202,341,343],{"href":342},"\u002Fblog\u002Fwhat-is-ai-governance\u002F","governance",". Governance does not replace routing. You need both.",[189,346,347],{},"Red flags: “we support many providers” with no task-class map; seat pricing that includes unlimited flagship; users pick the model in production; classify and memo share a model id in the demo; no quote before a run expands; fallback is “switch the dropdown”; graph does not record model id per step; forecasting done by an LLM in the demo on a short series.",[218,349,351],{"id":350},"what-to-look-for","What to look for",[223,353,354,360,366,376,382],{},[226,355,356,359],{},[192,357,358],{},"They can refuse the flagship."," Run a classify-only job and show the model id. If classify used the same model as the memo, routing is a slide. Refusal is the proof. Support for many models is not.",[226,361,362,365],{},[192,363,364],{},"Task-class map you can read."," Classify → compact. Retrieve → embeddings, not stuffing hundreds of tickets into a long window. Forecast → a time-series path, not a frontier model interpolating a spreadsheet. Reason → frontier. If they cannot name the classes, they cannot route them.",[226,367,368,371,372,375],{},[192,369,370],{},"Quotes before the run expands."," Operators see an estimate and can set a workstream cap. Seat licences hide routing. Unlimited flagship under a seat is always-flagship with a predictable opex line. See ",[202,373,374],{"href":252},"Total cost of ownership for enterprise AI",".",[226,377,378,381],{},[192,379,380],{},"Fallback is a logged promotion",", not “users will switch the dropdown.” Compact models fail on novel schemas and policy-edge language. Temporarily raise that class, budget-aware, then revert. A promotion without a log is a silent cost change. A dropdown is a training problem.",[226,383,384,387,388,375],{},[192,385,386],{},"The graph records the model id per step."," Six months later you can answer “which model drafted this?” without grepping provider dashboards. That is audit as well as cost. See ",[202,389,391],{"href":390},"\u002Fblog\u002Fwhat-is-a-lifecycle-graph\u002F","What is a lifecycle graph",[189,393,394,395,398,399,401],{},"Ask for four artefacts from one workstream run: model id per step; cost per task class; a classify job that did ",[262,396,397],{},"not"," use the flagship; a forecast that did ",[262,400,397],{}," use an LLM as the estimator. If the vendor can only show a chat transcript and a blended token total, routing is not in the product.",[189,403,404],{},"Why each artefact matters: model id is Measure in NIST language. Cost per class is how Finance splits labelling from reasoning. Classify-without-flagship is the refusal test. Forecast-without-LLM is whether they know the difference between narration and estimation. Demos are short series. Production is seasonality, holidays, and missing days.",[406,407,409],"h3",{"id":408},"what-a-live-demo-should-prove","What a live demo should prove",[189,411,412],{},"Do not accept a provider logo wall.",[414,415,416,424,427,430,433,436,439],"ol",{},[226,417,418,419,423],{},"Run one ",[202,420,422],{"href":421},"\u002Fblog\u002Fwhat-is-an-ai-workstream\u002F","workstream"," with extract, classify, retrieve, and a memo.",[226,425,426],{},"Show model id per step. Classify and extract are compact. The memo may be frontier.",[226,428,429],{},"Show cost (or tokens) per task class, quoted before the run grows, with a cap on the workstream.",[226,431,432],{},"Force a compact failure on a novel schema. Show a logged promotion, then a revert — not a user switching a dropdown.",[226,434,435],{},"Show a forecast path that is not an LLM interpolating a sheet. Narration can still be frontier.",[226,437,438],{},"Query the graph: which model drafted this payload? Answer from the ledger, not from a provider console.",[226,440,441],{},"Confirm a named-role gate still applies regardless of which model drafted. Routing chooses the brain. Governance still decides the write.",[189,443,444],{},"If they pass by opening a playground and picking “best,” you evaluated a dropdown.",[218,446,448],{"id":447},"how-this-shows-up-in-nimbus","How this shows up in Nimbus",[189,450,451],{},"Nimbus treats routing as an operating decision tied to workstream steps: task type, sensitivity, and cost — not “best everywhere.” Compact models handle extract. Frontier models are reserved for synthesis. Spend is quoted per workstream and visible per step.",[189,453,454,455,458],{},"Release gates apply regardless of which model drafted the payload. Connectors stay read-only by default. Routing decides ",[262,456,457],{},"which brain"," reads them. Governance still decides whether anything writes.",[189,460,461,462,465,466,468],{},"See ",[202,463,44],{"href":464},"\u002Fmodels\u002F",". For the unit of account, ",[202,467,253],{"href":252},". Score the four artefacts above. The product claim is the policy, not the catalogue.",[218,470,472],{"id":471},"questions-people-actually-ask","Questions people actually ask",[406,474,476],{"id":475},"is-model-agnostic-the-same-as-routing","Is “model-agnostic” the same as routing?",[189,478,479,480,482],{},"No. Model-agnostic means more than one provider. Routing means it ",[192,481,264],{}," by task class, with a default that is cheap where cheap is correct. A hidden always-flagship default is lock-in with extra logos. Ask what happens if the operator never touches a dropdown. If the answer is flagship, you have your policy.",[406,484,486],{"id":485},"should-operators-ever-pick-a-model","Should operators ever pick a model?",[189,488,489],{},"Rarely, and as an override. Production operators should brief outcomes. If quality depends on each user knowing which model is good at JSON, you have staffed a routing department by accident. Overrides should be logged, budget-aware, and exceptional. A dropdown on every run is how always-flagship returns through the side door.",[406,491,493],{"id":492},"why-not-put-forecasting-in-the-llm-if-the-numbers-look-fine-in-the-demo","Why not put forecasting in the LLM if the numbers look fine in the demo?",[189,495,496],{},"Demos are short series. Production is seasonality, holidays, and missing days. Keep narration on the frontier model and estimation on a time-series path. A fluent number is not a control. The warehouse or the statistical path already owns the number. RAG plus a frontier model is for policy language, not for revenue by region.",[406,498,500],{"id":499},"what-if-legal-requires-a-single-approved-model-vendor","What if legal requires a single approved model vendor?",[189,502,503,504,507],{},"Routing still applies ",[192,505,506],{},"inside"," that vendor’s catalogue: compact vs frontier vs embedding. Single-vendor is a contracting constraint, not an excuse to max tokens. Model-agnostic is nice. Task-class mapping inside one catalogue is the control. Do not skip routing because the RFP named one lab.",[406,509,511],{"id":510},"does-nis2-or-dora-change-the-routing-question","Does NIS2 or DORA change the routing question?",[189,513,514,515,518],{},"They change the ",[192,516,517],{},"evidence"," question. DORA and NIS2 expect you to understand ICT dependencies. “We are not sure which model ran last Tuesday” is a dependency you cannot explain. Record model id per step on the graph. That is enough to start. You do not need a new product category. You need Measure.",[218,520,522],{"id":521},"related-reading","Related reading",[189,524,525,528,529,533,534,375],{},[202,526,527],{"href":421},"What is an AI workstream",", ",[202,530,532],{"href":531},"\u002Fblog\u002Fhow-to-evaluate-an-agent-harness\u002F","how to evaluate an agent harness",", and ",[202,535,536],{"href":252},"what is AI token economics",[218,538,540],{"id":539},"sources","Sources",[223,542,543,548,553,558,563,568,573],{},[226,544,545],{},[202,546,547],{"href":204},"OpenAI API pricing",[226,549,550],{},[202,551,552],{"href":209},"Anthropic pricing",[226,554,555],{},[202,556,557],{"href":214},"Stanford HAI, 2025 AI Index Report",[226,559,560],{},[202,561,562],{"href":314},"McKinsey, The state of AI in 2025",[226,564,565],{},[202,566,567],{"href":321},"Gartner, AI TRiSM \u002F AI governance",[226,569,570],{},[202,571,572],{"href":330},"DORA (Regulation 2022\u002F2554)",[226,574,575],{},[202,576,577],{"href":334},"NIS2 (Directive 2022\u002F2555)",{"title":169,"searchDepth":170,"depth":170,"links":579},[580,581,582,586,587,594,595],{"id":220,"depth":170,"text":221},{"id":285,"depth":170,"text":286},{"id":350,"depth":170,"text":351,"children":583},[584],{"id":408,"depth":585,"text":409},3,{"id":447,"depth":170,"text":448},{"id":471,"depth":170,"text":472,"children":588},[589,590,591,592,593],{"id":475,"depth":585,"text":476},{"id":485,"depth":585,"text":486},{"id":492,"depth":585,"text":493},{"id":499,"depth":585,"text":500},{"id":510,"depth":585,"text":511},{"id":521,"depth":170,"text":522},{"id":539,"depth":170,"text":540},"2026-08-17","Model routing is a policy that uses a cheaper model for simple steps and a stronger model only when the task needs it — not a dropdown labelled “best.”",{"eyebrow":599,"title":600},"Short answers","What to check before you trust the router",[602,605,608],{"question":603,"answer":604},"What should a routing policy specify?","Which steps get a cheaper model, which steps require a stronger one, and the cost ceiling for the job.",{"question":606,"answer":607},"What is a warning sign?","A single default labelled best, with no record of which model actually ran.",{"question":609,"answer":610},"How do you know routing is working?","Simple steps cost less, hard steps still pass a check, and finance can see the split.","\u002Fblog\u002Fwhat-to-look-for-in-model-routing",{"title":179,"description":597},"evaluation","blog\u002Fwhat-to-look-for-in-model-routing",[613,616,617,618],"models","routing","cost","S2koGZz5yT1ldL2TBYZ9zTTxEZHhrGXowtqnJjeGBHw",{"hero":621,"id":623,"title":624,"archived":163,"authors":164,"badge":164,"body":625,"date":164,"definedTerm":164,"department":164,"description":629,"extension":172,"eyebrow":630,"faqHeader":164,"faqs":164,"footerBand":631,"headline":164,"image":164,"industry":164,"jobType":164,"listed":131,"location":164,"navigation":131,"openRoles":164,"pageLayout":164,"path":60,"relatedHeading":637,"seo":638,"series":164,"sitemap":131,"status":164,"stem":639,"subhead":164,"tags":164,"video":164,"whyJoin":164,"workplaceType":164,"__hash__":640},{"filename":622},"u2221455217_Flat_design_of_a_futuristic_minimalist_landscape__5d589295-cdea-4ea9-a262-be766881accf_1.png","content\u002Fblog\u002Findex.md","Exploring the future of intelligence.",{"type":166,"value":626,"toc":627},[],{"title":169,"searchDepth":170,"depth":170,"links":628},[],"Deep dives into pre-cognitive intelligence, sentient enterprises, and the evolving landscape of AI-driven business transformation.","Latest Research",{"headline":632,"description":633,"primaryLabel":634,"primaryTo":635,"secondaryLabel":636,"secondaryTo":12},"Stay at the frontier.","Subscribe for product updates and new insights.","Subscribe","\u002Fnewsletter","Explore the platform","More research",{"title":624,"description":629},"blog\u002Findex","BFSWGYO9bcTlaulivKYWyg08_DJHsdGg3OC6g_CG1Hw",[642,911],{"id":643,"title":644,"archived":163,"authors":645,"badge":647,"body":649,"date":891,"definedTerm":892,"department":164,"description":893,"extension":172,"eyebrow":164,"faqHeader":894,"faqs":896,"footerBand":164,"headline":164,"image":164,"industry":164,"jobType":164,"listed":131,"location":164,"navigation":131,"openRoles":164,"pageLayout":164,"path":905,"relatedHeading":164,"seo":906,"series":907,"sitemap":131,"status":164,"stem":908,"subhead":164,"tags":909,"video":164,"whyJoin":164,"workplaceType":164,"__hash__":910},"content\u002Fblog\u002Fwhat-is-collaborative-ai.md","What is collaborative AI, and why should you care?",[646],{"name":182,"to":183},{"label":648},"Use case",{"type":166,"value":650,"toc":883},[651,654,661,665,668,682,685,692,704,708,711,721,727,764,768,771,779,786,789,793,796,810,826,833,837,840,852,859,863,866,874],[189,652,653],{},"Collaborative AI is when several people work on one piece of work with the same context, the same tools, and a named finish line — and the AI is in that shared place, not only in each person’s private chat.",[189,655,656,657,660],{},"Personal assistants are good at drafting. Shared jobs need a shared room. The ",[202,658,100],{"href":659},"\u002Fuse-cases\u002F"," hub is the rest of the series.",[218,662,664],{"id":663},"what-is-collaborative-ai","What is collaborative AI?",[189,666,667],{},"At minimum it has four parts:",[223,669,670,673,676,679],{},[226,671,672],{},"A named job (not “the channel”).",[226,674,675],{},"People who can see the same files and history.",[226,677,678],{},"Tools that read, and sometimes write, with a recorded step.",[226,680,681],{},"Someone who can say the change does not go out.",[189,683,684],{},"That is different from giving the team one login to a chatbot. The login is still a personal product. Collaborative AI is the job as the unit.",[189,686,687,688,691],{},"McKinsey’s ",[202,689,690],{"href":314},"State of AI"," (2025) found that 88% of organisations use AI in at least one function, while most are still in the pilot stage. Use is common. Shared process is not.",[189,693,694,698,699,703],{},[202,695,697],{"href":696},"\u002Fblog\u002Fmultiplayer-ai-and-multi-agent-ai\u002F","Multiplayer AI vs multi-agent AI"," is the sibling distinction for several people in one session versus a cast of models. ",[202,700,702],{"href":701},"\u002Fblog\u002Fcollaborative-ai-and-personal-assistants\u002F","Collaborative AI and personal assistants"," is the comparison with copilots.",[218,705,707],{"id":706},"is-collaborative-ai-the-same-as-multiplayer-ai","Is collaborative AI the same as multiplayer AI?",[189,709,710],{},"They overlap. They are not the same purchase question.",[189,712,713,716,717,720],{},[192,714,715],{},"Multiplayer AI"," stresses the live room: can a second department join, see the same brief, and reject a proposal while the work is happening. The ",[202,718,719],{"href":696},"multiplayer AI"," page is that definition.",[189,722,723,726],{},[192,724,725],{},"Collaborative AI"," stresses the shared job, including after everyone hangs up. The files, the named stop, and the outcome still have a home on Monday. A huddle that produced a fluent recap and no signed artefact was multiplayer for an hour and not collaborative.",[189,728,729,730,732,733,737,738,528,742,528,746,750,751,528,755,533,759,763],{},"You can have multiplayer without collaboration: five people in a call, nothing written down. You can have collaboration without a live session: a ",[202,731,422],{"href":421}," that two teams open on different days. Most cross-department work that writes to a live system needs both. ",[202,734,736],{"href":735},"\u002Fblog\u002Fhow-to-evaluate-collaborative-ai\u002F","How to evaluate collaborative AI"," is the sheet for the job. Function walkthroughs: ",[202,739,741],{"href":740},"\u002Fblog\u002Fcollaborative-ai-for-operations\u002F","operations",[202,743,745],{"href":744},"\u002Fblog\u002Fcollaborative-ai-for-human-resources\u002F","human resources",[202,747,749],{"href":748},"\u002Fblog\u002Fcollaborative-ai-for-customer-support\u002F","customer support",", and the existing ",[202,752,754],{"href":753},"\u002Fblog\u002Fcollaborative-ai-for-revenue-operations\u002F","revenue",[202,756,758],{"href":757},"\u002Fblog\u002Fcollaborative-ai-for-finance-and-planning\u002F","finance",[202,760,762],{"href":761},"\u002Fblog\u002Fcollaborative-ai-for-legal-and-compliance-review\u002F","legal"," guides.",[218,765,767],{"id":766},"why-should-you-care-about-collaborative-ai","Why should you care about collaborative AI?",[189,769,770],{},"You should care when a mistake is expensive because two teams thought they were looking at the same thing and were not.",[189,772,773,774,778],{},"Yang and colleagues, writing in ",[202,775,777],{"href":776},"https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41562-021-01196-4","Nature Human Behaviour"," (2022), found that firm-wide remote work made collaboration networks more static and siloed. Shared jobs already fight that pull. AI that lives only in private threads can make the silo worse: each person has a fluent answer, and nobody has the same file.",[189,780,781,785],{},[202,782,784],{"href":783},"https:\u002F\u002Fwww.melconway.com\u002FHome\u002FCommittees_Paper.html","Melvin Conway’s 1968 paper"," noted that organisations design systems that copy their communication structure. If the structure is one person and one chat, the AI will copy that. Collaborative AI is a chance to copy the meeting you already needed, not the inbox.",[189,787,788],{},"If your work is mostly solo drafting, a personal assistant may be the right tool. Collaborative AI is for the jobs that already required a meeting.",[218,790,792],{"id":791},"how-do-you-get-more-from-collaborative-ai","How do you get more from collaborative AI?",[189,794,795],{},"Treat it like a project room, not like a better chatbot.",[414,797,798,801,804,807],{},[226,799,800],{},"Name the job and the finish line before you invite the model.",[226,802,803],{},"Put the files in the room, not in five inboxes.",[226,805,806],{},"Write down who can approve a change to a live system (when AI changes a customer record, a forecast, or a ticket).",[226,808,809],{},"Keep the debate in chat if you like; keep the outcome in the room.",[189,811,812,815,816,820,821,825],{},[202,813,814],{"href":421},"What an AI workstream is"," is the container for step two. ",[202,817,819],{"href":818},"\u002Fblog\u002Frbac-for-enterprise-ai\u002F","RBAC for enterprise AI"," is step three. ",[202,822,824],{"href":823},"\u002Fblog\u002Fwhat-is-harness-engineering\u002F","Harness engineering"," is why the prompt alone is not the system.",[189,827,828,829,832],{},"If you are asking how the model is boxed in — tools, stops, checks — that is ",[202,830,831],{"href":823},"harness engineering",", the practice of improving the environment around the model rather than only the wording of the ask.",[218,834,836],{"id":835},"what-does-this-look-like-on-a-real-job","What does this look like on a real job?",[189,838,839],{},"Sales wants a discount exception. Finance wants the margin intact. A personal assistant can draft the email. Collaborative AI would be a shared workstream: the CRM excerpt, the margin sheet, both teams in the same history, and a recorded approve before anything writes back to the account — write-back meaning AI changes a live system.",[189,841,842,843,528,846,533,849,375],{},"Legal reviewing a clause with operations in the same place is the same pattern. So is a forecast that planning and FP&A both own. We walk those through in ",[202,844,845],{"href":753},"collaborative AI for revenue operations",[202,847,848],{"href":757},"finance and planning",[202,850,851],{"href":761},"legal and compliance review",[189,853,854,855,858],{},"If two teams still disagree after the files are in the room, that is expected. ",[202,856,857],{"href":761},"Collaborative AI for legal and compliance review"," is how to keep the quoted clause from pretending the argument is over.",[218,860,862],{"id":861},"how-do-you-start-without-a-big-programme","How do you start without a big programme?",[189,864,865],{},"Pick one recurring cross-team job. Create one shared place. Add the people who already argue about it. Attach the two files they always forward. Decide who can say no. Run it for two cycles. Measure whether you still paste the same screenshot into chat.",[189,867,868,869,873],{},"Nimbus’s ",[202,870,872],{"href":871},"\u002Fproduct\u002Fworkstreams\u002F","workstreams"," are built for that shape. You can try the same shape in a wiki plus a ticket if that is what you have.",[189,875,876,877,880,881,375],{},"For how work should be handed between teams, see ",[202,878,879],{"href":757},"collaborative AI for finance and planning",". For who is allowed to see the room, start with ",[202,882,819],{"href":818},{"title":169,"searchDepth":170,"depth":170,"links":884},[885,886,887,888,889,890],{"id":663,"depth":170,"text":664},{"id":706,"depth":170,"text":707},{"id":766,"depth":170,"text":767},{"id":791,"depth":170,"text":792},{"id":835,"depth":170,"text":836},{"id":861,"depth":170,"text":862},"2026-08-14","collaborative AI","Collaborative AI is several people sharing one job — context, tools, and a finish line — rather than each person using their own assistant.",{"eyebrow":599,"title":895},"Questions about the definition",[897,900,903],{"question":898,"answer":899},"Is collaborative AI the same as a shared ChatGPT login?","No. A shared login is still a personal assistant used by several people. Collaborative AI is one job with shared context, shared tools, and a named person who can stop a change.",{"question":901,"answer":902},"Do we need it if everyone already has a copilot?","You need it when the work crosses departments or writes to a live system. A copilot helps one person draft. Collaborative AI is the place the draft becomes a shared result.",{"question":707,"answer":904},"They overlap. Multiplayer AI stresses several people in one session. Collaborative AI stresses the shared job — including what happens after the session ends.","\u002Fblog\u002Fwhat-is-collaborative-ai",{"title":644,"description":893},"collaborative-ai","blog\u002Fwhat-is-collaborative-ai",[907,892],"I5ePt2tkJTa2ss64T-c7vERntTA_0xBPngJUWlqoLh0",{"id":912,"title":913,"archived":163,"authors":914,"badge":916,"body":918,"date":596,"definedTerm":164,"department":164,"description":922,"extension":172,"eyebrow":164,"faqHeader":1391,"faqs":1393,"footerBand":164,"headline":164,"image":164,"industry":164,"jobType":164,"listed":131,"location":164,"navigation":131,"openRoles":164,"pageLayout":164,"path":1403,"relatedHeading":164,"seo":1404,"series":1405,"sitemap":131,"status":164,"stem":1406,"subhead":164,"tags":1407,"video":164,"whyJoin":164,"workplaceType":164,"__hash__":1410},"content\u002Fblog\u002Fwhat-is-an-enterprise-ai-operating-system.md","What is an Enterprise AI Operating System",[915],{"name":182,"to":183},{"label":917},"Explainer",{"type":166,"value":919,"toc":1367},[920,923,926,929,934,937,939,1006,1009,1041,1053,1060,1062,1068,1071,1074,1077,1080,1084,1090,1103,1109,1115,1121,1125,1131,1145,1151,1157,1163,1166,1175,1177,1180,1221,1240,1242,1246,1249,1253,1261,1265,1268,1272,1275,1279,1282,1286,1289,1293,1296,1300,1307,1311,1314,1318,1323,1327,1330,1334,1345,1347,1353,1355],[189,921,922],{},"An enterprise AI operating system is the layer between the AI model and how departments actually work — like Windows sits between the chip and your apps.",[189,924,925],{},"If your question is “which model should we buy,” you are shopping for a chip. If your question is “how do revenue, legal, and finance run the same loop without a personal-account workaround,” you are shopping for an OS.",[189,927,928],{},"It is not a chatbot with company login. It is not a model API with a prompt library.",[189,930,931,933],{},[202,932,315],{"href":314}," found that 88% of companies use AI in at least one function — and that a majority are still piloting. About one in three report that they are scaling. Buying another model does not close that gap. The missing layer is how work actually runs.",[189,935,936],{},"The OS metaphor is useful if you keep it honest. An operating system does not replace your spreadsheet or your CRM. It gives applications isolation, permissions, input and output, and a place to keep state after the window closes. An enterprise AI OS does the same for work that uses models: isolation of jobs, rights over tools and data, reads and writes to live systems, budgets, and a record that survives the session.",[218,938,221],{"id":220},[223,940,941,947,953,963,969,980,988,998],{},[226,942,943,946],{},[192,944,945],{},"Copilot."," A high-quality assistant for a person. Admin controls, company login. Not, by itself, how several departments finish one job under a named signer. At work, this is “help me draft.” It is not “release this CRM change.”",[226,948,949,952],{},[192,950,951],{},"Operating system (in this sense)."," Process isolation, permissions, input\u002Foutput to live systems, budgeting, and durable state — the jobs a kernel does for apps.",[226,954,955,958,959,962],{},[192,956,957],{},"System of record."," CRM, ERP, HR — still authoritative. The OS is the system of ",[262,960,961],{},"work",", not a second CRM.",[226,964,965,968],{},[192,966,967],{},"Forward-deployed engineer."," A vendor consultant who sits with you for months. Some programmes need that. Many companies need governed work this quarter without it.",[226,970,971,974,975,979],{},[192,972,973],{},"NIST AI RMF."," ",[202,976,978],{"href":977},"https:\u002F\u002Fwww.nist.gov\u002Fitl\u002Fai-risk-management-framework","Govern, Map, Measure, Manage"," — public-sector language for the same kernel idea: identity, tool rights, and a record attached to real actions.",[226,981,982,985,986,375],{},[192,983,984],{},"Workstream."," The process-isolation unit: one job, one scope, one finish line. See ",[202,987,527],{"href":421},[226,989,990,993,994,375],{},[192,991,992],{},"Fail-closed writes."," Missing named signer means nothing happens. See ",[202,995,997],{"href":996},"\u002Fblog\u002Fwhat-is-write-back-governance\u002F","What is write-back governance",[226,999,1000,1003,1004,375],{},[192,1001,1002],{},"Usage limits and quotes."," Work is quoted before it runs and capped by limits and budgets. See ",[202,1005,253],{"href":252},[189,1007,1008],{},"Five jobs cluster around the term:",[414,1010,1011,1017,1023,1029,1035],{},[226,1012,1013,1016],{},[192,1014,1015],{},"Process isolation."," Go-to-market does not silently inherit finance’s ERP login.",[226,1018,1019,1022],{},[192,1020,1021],{},"Resource management."," Inference and tool calls are budgeted.",[226,1024,1025,1028],{},[192,1026,1027],{},"I\u002FO control."," Reads and writes to CRM and ERP are first-class — not “chat that sometimes calls an API.”",[226,1030,1031,1034],{},[192,1032,1033],{},"Permissioning."," Identity and context decide what an agent can see and do. A signed PDF is not enforcement.",[226,1036,1037,1040],{},[192,1038,1039],{},"Durable state."," Outcomes, approvals, and rationale survive the session.",[189,1042,1043,1044,206,1048,1052],{},"Copilots generally fail the last three. ",[202,1045,1047],{"href":1046},"\u002Fblog\u002Fnimbus-vs-chatgpt-enterprise\u002F","ChatGPT Enterprise",[202,1049,1051],{"href":1050},"\u002Fblog\u002Fnimbus-vs-claude\u002F","Claude for Work"," are excellent assistants. They are not this job.",[189,1054,1055,1059],{},[202,1056,1058],{"href":1057},"\u002Fblog\u002Fwhat-is-model-context-protocol\u002F","Model Context Protocol"," is also not this job. A common plug for tools is USB. USB did not create Windows.",[218,1061,286],{"id":285},[189,1063,1064,1065,1067],{},"Operators do not “open the OS” the way they open a model playground. They open ",[192,1066,961],{},": a brief, a scoped live system, a review, a release.",[189,1069,1070],{},"It affects you if AI is starting to touch revenue, financial close, customer records, or regulated processes. Chat history does not answer “who approved this, against which policy?”",[189,1072,1073],{},"The OS also matters if you refuse a six-to-twelve-month vendor-engineer programme as the only path to production.",[189,1075,1076],{},"The anti-pattern is using an OS as a better chatbot: one user, one thread, no write path, no memory beyond the conversation. If nobody except the original operator can reconstruct what happened, you have a log, not an operating system.",[189,1078,1079],{},"Personal copilots optimise for “the model always answers.” An OS optimises for “the company only acts when the gate says so.” A spend cap or a missing approval is a successful outcome.",[406,1081,1083],{"id":1082},"what-changes-by-role","What changes by role",[189,1085,1086,1089],{},[192,1087,1088],{},"Finance."," The OS is how close and forecast jobs get a budget, a read-only ERP connector, a wiki checklist, and a named signer — without a second ledger. Finance should still own NetSuite. The OS should point at it.",[189,1091,1092,1095,1096,1099,1100,375],{},[192,1093,1094],{},"Legal."," Reconstructable authorisation, purpose-limited scope, and a place that is not a personal chat vendor. Legal should evaluate whether unapproved writes are ",[262,1097,1098],{},"impossible",", not whether a policy PDF exists. See ",[202,1101,1102],{"href":342},"What is AI governance",[189,1104,1105,1108],{},[192,1106,1107],{},"Operations."," Isolation and durable state are ops problems. Ops should ask whether a paused run is a first-class object, whether connectors default to read-only, and whether Perception (or equivalent) can answer “why did this change?” without a data team reconstructing Slack.",[189,1110,1111,1114],{},[192,1112,1113],{},"Go-to-market."," Cross-department loops — legal on a renewal, finance on a discount — need a shared job, not a shared inbox. GTM should not have to choose between a copilot that cannot write safely and a spreadsheet export to a consumer model.",[189,1116,1117,1120],{},[192,1118,1119],{},"Security."," Identity, least privilege, fail-closed I\u002FO, and not turning the OS into a second store of the whole company. Security also cares that self-service configuration does not mean tenant-wide write keys.",[406,1122,1124],{"id":1123},"what-people-get-wrong","What people get wrong",[189,1126,1127,1130],{},[192,1128,1129],{},"“ChatGPT with integrations.”"," Plugins without scoped work, approval architecture, and durable decision records are plugins. A copilot with automation actions can move data. It cannot, by itself, make unapproved writes impossible.",[189,1132,1133,1136,1137,1140,1141,1144],{},[192,1134,1135],{},"MLOps as a substitute."," MLOps governs ",[262,1138,1139],{},"model production",". An enterprise AI OS governs ",[262,1142,1143],{},"operational work that uses models",". They stack.",[189,1146,1147,1150],{},[192,1148,1149],{},"Replacing the CRM."," Salesforce, NetSuite, Workday, and the warehouse remain authoritative. Duplicating them is a second system of record.",[189,1152,1153,1156],{},[192,1154,1155],{},"OS as chatbot."," One user, one thread, no write path, no memory. That is a copilot with extra vocabulary.",[189,1158,1159,1162],{},[192,1160,1161],{},"Forward-deployed as the only path."," Some warehouses need specialists. Most operators need to attach a connector and set a named signer in the UI.",[189,1164,1165],{},"Good looks like: workstreams, wiki, read-only-default connectors, agent teams, Lifecycle Graph, model routing, run quotes, fail-closed writes, self-service configuration. Failure looks like another model contract plus a six-month SOW.",[189,1167,1168,1169,1172,1173,375],{},"For the copilot-versus-OS choice, see ",[202,1170,1171],{"href":701},"collaborative AI and personal assistants",". For vendor scoring, ",[202,1174,532],{"href":531},[218,1176,448],{"id":447},[189,1178,1179],{},"Nimbus is a self-service enterprise AI OS. Operators configure it in the product.",[223,1181,1182,1189,1195,1201,1207,1216],{},[226,1183,1184,1188],{},[192,1185,1186],{},[202,1187,31],{"href":421}," isolate process.",[226,1190,1191,1194],{},[192,1192,1193],{},"Wiki"," holds asserted policy — approved playbooks, not a dump of PDFs a search might find.",[226,1196,1197,1200],{},[192,1198,1199],{},"Connectors"," attach live systems. Default is read-only. Write-back is opt-in and gated.",[226,1202,1203,1206],{},[192,1204,1205],{},"Agent teams"," are department-shaped.",[226,1208,1209,1211,1212,1215],{},[192,1210,23],{}," stores the causal record. ",[192,1213,1214],{},"Perception"," queries it in ordinary language.",[226,1217,1218,1220],{},[192,1219,194],{}," puts routine extract on cheaper models.",[189,1222,461,1223,206,1226,1228,1229,528,1231,528,1234,528,1237,375],{},[202,1224,11],{"href":1225},"\u002Foverview\u002F",[202,1227,532],{"href":531},". Product surfaces: ",[202,1230,31],{"href":871},[202,1232,39],{"href":1233},"\u002Fproduct\u002Fgovernance\u002F",[202,1235,23],{"href":1236},"\u002Fproduct\u002Flifecycle-graph\u002F",[202,1238,1214],{"href":1239},"\u002Fproduct\u002Fperception\u002F",[218,1241,472],{"id":471},[406,1243,1245],{"id":1244},"is-an-enterprise-ai-os-just-chatgpt-with-integrations","Is an enterprise AI OS just “ChatGPT with integrations”?",[189,1247,1248],{},"No. Integrations without scoped work, approval architecture, and durable decision records are plugins. A copilot with automation actions can move data. It cannot, by itself, make unapproved writes impossible.",[406,1250,1252],{"id":1251},"how-is-this-different-from-mlops","How is this different from MLOps?",[189,1254,1255,1256,1140,1258,1260],{},"MLOps governs ",[262,1257,1139],{},[262,1259,1143],{},". They stack. They do not substitute.",[406,1262,1264],{"id":1263},"do-we-still-need-a-crm-if-we-buy-an-os","Do we still need a CRM if we buy an OS?",[189,1266,1267],{},"Yes. Salesforce, NetSuite, Workday, and the warehouse remain authoritative.",[406,1269,1271],{"id":1270},"does-every-company-need-an-os","Does every company need an OS?",[189,1273,1274],{},"If the job is personal drafting with no writes to live systems, a governed copilot may be enough. The OS becomes the right abstraction when work crosses departments, when writes are material, and when you must reconstruct decisions.",[406,1276,1278],{"id":1277},"is-this-the-same-as-an-integration-platform-ipaas","Is this the same as an integration platform (iPaaS)?",[189,1280,1281],{},"No. iPaaS moves data on schedules and triggers. An AI OS runs language-using jobs with scope, spend, and a human gate. You may still need iPaaS. It does not quote a named signer on a CRM payload.",[406,1283,1285],{"id":1284},"does-operating-system-mean-we-install-software-on-laptops","Does “operating system” mean we install software on laptops?",[189,1287,1288],{},"No. It is a layer for work, not a desktop kernel. The metaphor is isolation, permissions, I\u002FO, and state.",[406,1290,1292],{"id":1291},"can-we-build-this-ourselves-on-a-model-api","Can we build this ourselves on a model API?",[189,1294,1295],{},"You can assemble pieces. You will still need isolation, connectors, gates, spend, and a graph. Most “we built a GPT” programmes stall at the copilot layer. McKinsey’s split between using AI and scaling it is that stall in survey form.",[406,1297,1299],{"id":1298},"where-do-agent-teams-fit","Where do agent teams fit?",[189,1301,1302,1303,375],{},"They are the department-shaped specialists the OS schedules onto workstreams. They are not the OS. See ",[202,1304,1306],{"href":1305},"\u002Fblog\u002Fwhat-is-multi-agent-ai\u002F","What is multi-agent AI",[406,1308,1310],{"id":1309},"how-does-nists-ai-rmf-map","How does NIST’s AI RMF map?",[189,1312,1313],{},"Govern (owners, policy), Map (inventory of jobs and systems), Measure (evidence, spend, rejects), Manage (fail-closed writes, incident path). A product can make those cheaper. A framework PDF cannot enforce them.",[406,1315,1317],{"id":1316},"what-is-perception-in-this-picture","What is Perception in this picture?",[189,1319,1320,1321,375],{},"Ordinary-language questions over the company’s graph, wiki, and scoped systems — with the next step being a workstream, not another search. See ",[202,1322,1214],{"href":1239},[406,1324,1326],{"id":1325},"do-we-need-a-forward-deployed-engineer-to-go-live","Do we need a forward-deployed engineer to go live?",[189,1328,1329],{},"Nothe default path. If operators cannot attach a read-only connector and set a named signer in the UI, you do not have a self-service OS. Specialists belong on genuine exceptions, such as a warehouse with no OAuth.",[406,1331,1333],{"id":1332},"is-search-rag-an-os","Is search (RAG) an OS?",[189,1335,1336,1337,206,1341,375],{},"No. Lookup-then-answer is infrastructure. It does not isolate jobs or gate writes. See ",[202,1338,1340],{"href":1339},"\u002Fblog\u002Fwhat-is-enterprise-rag\u002F","What is enterprise RAG",[202,1342,1344],{"href":1343},"\u002Fblog\u002Fnimbus-vs-glean\u002F","Nimbus vs Glean",[218,1346,522],{"id":521},[189,1348,1349,206,1351,375],{},[202,1350,527],{"href":421},[202,1352,1102],{"href":342},[218,1354,540],{"id":539},[223,1356,1357,1362],{},[226,1358,1359],{},[202,1360,1361],{"href":314},"McKinsey, The state of AI (2025)",[226,1363,1364],{},[202,1365,1366],{"href":977},"NIST AI Risk Management Framework",{"title":169,"searchDepth":170,"depth":170,"links":1368},[1369,1370,1374,1375,1389,1390],{"id":220,"depth":170,"text":221},{"id":285,"depth":170,"text":286,"children":1371},[1372,1373],{"id":1082,"depth":585,"text":1083},{"id":1123,"depth":585,"text":1124},{"id":447,"depth":170,"text":448},{"id":471,"depth":170,"text":472,"children":1376},[1377,1378,1379,1380,1381,1382,1383,1384,1385,1386,1387,1388],{"id":1244,"depth":585,"text":1245},{"id":1251,"depth":585,"text":1252},{"id":1263,"depth":585,"text":1264},{"id":1270,"depth":585,"text":1271},{"id":1277,"depth":585,"text":1278},{"id":1284,"depth":585,"text":1285},{"id":1291,"depth":585,"text":1292},{"id":1298,"depth":585,"text":1299},{"id":1309,"depth":585,"text":1310},{"id":1316,"depth":585,"text":1317},{"id":1325,"depth":585,"text":1326},{"id":1332,"depth":585,"text":1333},{"id":521,"depth":170,"text":522},{"id":539,"depth":170,"text":540},{"eyebrow":599,"title":1392},"The layer between the model and the work",[1394,1397,1400],{"question":1395,"answer":1396},"What is an enterprise AI operating system?","The layer between the AI model and how departments actually work, the way an operating system sits between the chip and the applications.",{"question":1398,"answer":1399},"Is the model the operating system?","No. The model is a component. The operating system is how work, permissions, and records are organised around it.",{"question":1401,"answer":1402},"What should departments find there?","The same job, the same rules, and a record, instead of a different chatbot in every function.","\u002Fblog\u002Fwhat-is-an-enterprise-ai-operating-system",{"title":913,"description":922},"explainer","blog\u002Fwhat-is-an-enterprise-ai-operating-system",[1405,1408,1409,343],"enterprise-ai","operating-system","viSXGpdmqdd2fuaUmfH6OngdOgRbSEtUnbseaPgQvis",{"enabled":163,"message":1412,"linkLabel":79,"linkHref":80,"id":1413,"title":1414,"archived":163,"authors":164,"badge":164,"body":1415,"date":164,"definedTerm":164,"department":164,"description":169,"extension":172,"eyebrow":164,"faqHeader":164,"faqs":164,"footerBand":164,"headline":164,"image":164,"industry":164,"jobType":164,"listed":131,"location":164,"navigation":131,"openRoles":164,"pageLayout":164,"path":1419,"relatedHeading":164,"seo":1420,"series":164,"sitemap":163,"status":164,"stem":1421,"subhead":164,"tags":164,"video":164,"whyJoin":164,"workplaceType":164,"__hash__":1422},"We're hiring! Join the team building the Sentient Enterprise.","content\u002Fshared\u002Fhiring.md","Hiring banner",{"type":166,"value":1416,"toc":1417},[],{"title":169,"searchDepth":170,"depth":170,"links":1418},[],"\u002Fshared\u002Fhiring",{"title":1414,"description":169},"shared\u002Fhiring","1zs3boivKda1e-b-hAyuNcmZSKjZUAXmecnwHVgcHzk",{"fold":1424,"id":1428,"title":1429,"archived":163,"authors":164,"badge":164,"body":1430,"date":164,"definedTerm":164,"department":164,"description":169,"extension":172,"eyebrow":164,"faqHeader":164,"faqs":164,"footerBand":1434,"headline":164,"image":164,"industry":164,"jobType":164,"listed":131,"location":164,"navigation":131,"openRoles":164,"pageLayout":164,"path":1438,"relatedHeading":164,"seo":1439,"series":164,"sitemap":163,"status":164,"stem":1440,"subhead":164,"tags":164,"video":164,"whyJoin":164,"workplaceType":164,"__hash__":1441},{"headline":1425,"description":1426,"primaryLabel":8,"primaryTo":1427,"secondaryLabel":636,"secondaryTo":12},"Run frontier AI your business actually owns.","Governed workstreams, 3,000+ integrations, and a proprietary knowledge graph. Start on Free.","\u002Fsignup?plan=free","content\u002Fshared\u002Fcta.md","Site CTAs",{"type":166,"value":1431,"toc":1432},[],{"title":169,"searchDepth":170,"depth":170,"links":1433},[],{"headline":1435,"description":1436,"primaryLabel":8,"primaryTo":1427,"secondaryLabel":1437,"secondaryTo":85},"See what governed AI looks like on your stack.","Connect your tools, run a workstream, and keep every decision on your ledger. Start on Free.","Talk to our team","\u002Fshared\u002Fcta",{"title":1429,"description":169},"shared\u002Fcta","eYqahyaPnbp8GKrWpoORbZdtmkWmgHr5F61ZHOnb8sY",1792999485848]