[{"data":1,"prerenderedAt":1756},["ShallowReactive",2],{"site-nav-content":3,"blog:/blog/what-is-ai-governance":163,"blog-index-copy":782,"blog:/blog/what-is-ai-governance:surround":803,"hiring-banner-content":1725,"site-cta-content":1737},{"header":4,"productNav":9,"nav":42,"footer":61,"askAI":116,"id":147,"title":148,"archived":149,"authors":150,"badge":150,"body":151,"date":150,"department":150,"description":155,"extension":158,"eyebrow":150,"faqHeader":150,"faqs":150,"footerBand":150,"headline":150,"image":150,"industry":150,"jobType":150,"listed":149,"location":150,"navigation":115,"openRoles":150,"pageLayout":150,"path":159,"relatedHeading":150,"seo":160,"series":150,"sitemap":115,"status":150,"stem":161,"subhead":150,"tags":150,"video":150,"whyJoin":150,"workplaceType":150,"__hash__":162},{"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","/overview","Seven layers. One closed loop.",{"label":15,"to":16,"description":17},"Conflux","/product/conflux","Where your team, workstreams, and agents meet.",{"label":19,"to":20,"description":21},"Agent Teams","/product/agent-teams","Specialist teams - governed from day one.",{"label":23,"to":24,"description":25},"Lifecycle Graph","/product/lifecycle-graph","Intelligence that compounds across every interaction.",{"label":27,"to":28,"description":29},"Company Wiki","/product/wiki","Playbooks and policies where expertise stays.",{"label":31,"to":32,"description":33},"Workstreams","/product/workstreams","From brief to signed-off deliverable on one canvas.",{"label":35,"to":36,"description":37},"Perception Console","/product/perception","Ask your whole business in plain English.",{"label":39,"to":40,"description":41},"Governance","/product/governance","Frontier AI you can actually sign off on.",[43,46,49,52,55,58],{"label":44,"to":45},"Models","/models",{"label":47,"to":48},"Pricing","/pricing",{"label":50,"to":51},"Integrations","/integrations",{"label":53,"to":54},"Security","/security",{"label":56,"to":57},"Partners","/partners",{"label":59,"to":60},"Insights","/blog",{"productHeading":5,"companyHeading":62,"legalHeading":63,"docsLabel":64,"docsUrl":65,"statementLines":66,"copyright":69,"companyLinks":70,"legalLinks":85,"socialLinks":95,"bottomLinks":105},"Company","Legal","Docs","https://docs.gonimbus.ai",[67,68],"Stop training someone else's model.","Control your AI.","© 2026 Nimbus Intelligence, Inc. All rights reserved.",[71,72,73,74,75,77,80,83],{"label":47,"to":48},{"label":50,"to":51},{"label":53,"to":54},{"label":59,"to":60},{"label":76,"to":57},"Partner Program",{"label":78,"to":79},"Careers","/careers",{"label":81,"to":82},"System status","/status",{"label":7,"to":84},"/contact",[86,89,92],{"label":87,"to":88},"Terms of Service","/terms",{"label":90,"to":91},"Privacy Policy","/privacy",{"label":93,"to":94},"Compliance","/compliance",[96,99,102],{"label":97,"href":98},"LinkedIn","https://www.linkedin.com/company/gonimbusai/",{"label":100,"href":101},"X","https://x.com/gonimbusai",{"label":103,"href":104},"Instagram","https://www.instagram.com/gonimbus_ai/",[106,108,110,111,112],{"label":107,"to":88},"Terms",{"label":109,"to":91},"Privacy",{"label":93,"to":94},{"label":81,"to":82},{"label":113,"to":114,"external":115},"LLMs.txt","/llms.txt",true,{"text":117,"prompt":118},"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":119,"platforms":121},{" Summarize the highlights from Nimbus's website":120},"https://gonimbus.ai",[122,127,132,137,142],{"name":123,"label":124,"icon":125,"hrefPrefix":126},"chatgpt","ChatGPT","simple-icons:openai","https://chatgpt.com/?prompt=",{"name":128,"label":129,"icon":130,"hrefPrefix":131},"perplexity","Perplexity","mdi:magnify","https://www.perplexity.ai/search/new?q=",{"name":133,"label":134,"icon":135,"hrefPrefix":136},"grok","Grok","simple-icons:x","https://x.com/i/grok?text=",{"name":138,"label":139,"icon":140,"hrefPrefix":141},"claude","Claude","simple-icons:anthropic","https://claude.ai/new?q=",{"name":143,"label":144,"icon":145,"hrefPrefix":146},"google-ai","Google AI","simple-icons:google","https://www.google.com/search?udm=50&aep=11&q=","content/shared/nav.md","Site navigation",false,null,{"type":152,"value":153,"toc":154},"minimark",[],{"title":155,"searchDepth":156,"depth":156,"links":157},"",2,[],"md","/shared/nav",{"title":148,"description":155},"shared/nav","p6IsjEfjcrwkjwvGsPcuEkpoJSeUn2vJXPTxLWQfw9M",{"id":164,"title":165,"archived":149,"authors":166,"badge":169,"body":171,"date":771,"department":150,"description":772,"extension":158,"eyebrow":150,"faqHeader":150,"faqs":150,"footerBand":150,"headline":150,"image":150,"industry":150,"jobType":150,"listed":149,"location":150,"navigation":115,"openRoles":150,"pageLayout":150,"path":773,"relatedHeading":150,"seo":774,"series":775,"sitemap":115,"status":150,"stem":776,"subhead":150,"tags":777,"video":150,"whyJoin":150,"workplaceType":150,"__hash__":781},"content/blog/what-is-ai-governance.md","What is AI Governance",[167],{"name":168,"to":120},"Nimbus Research",{"label":170},"Explainer",{"type":152,"value":172,"toc":746},[173,182,185,188,209,212,226,231,332,342,346,355,358,390,393,423,430,435,441,453,459,465,477,481,487,493,499,507,513,521,540,544,547,554,568,578,582,586,594,598,601,605,612,616,619,623,626,630,637,641,644,648,651,655,658,662,665,669,674,678,681,685,696,700],[174,175,176,177,181],"p",{},"AI governance is the set of rules, ",[178,179,180],"strong",{},"enforced in the software people actually use",", that decide who may use which AI, on which company data, and whether that AI is allowed to change a live business system — plus a record of what happened afterwards.",[174,183,184],{},"A training video is not that. An acceptable-use PDF is not that. An admin toggle the model can ignore is not that. If an unapproved change can still succeed, you have guidance, not governance.",[174,186,187],{},"People use the phrase for three different things, and they get mixed up:",[189,190,191,203,206],"ol",{},[192,193,194,195,202],"li",{},"A public commitment — for example the ",[196,197,201],"a",{"href":198,"rel":199},"https://oecd.ai/en/ai-principles",[200],"nofollow","OECD AI Principles",".",[192,204,205],{},"A company committee with a risk register.",[192,207,208],{},"The runtime that actually stops a change to CRM, ERP, or a customer message.",[174,210,211],{},"All three are real. Only the third one would have blocked an unlogged field change that later showed up in a forecast.",[174,213,214,219,220,225],{},[196,215,218],{"href":216,"rel":217},"https://www.gartner.com/en/articles/ai-governance-trism",[200],"Gartner’s TRiSM"," language is about that third layer: trust, risk, and security around the systems that run — not a quarterly slide about principles. The ",[196,221,224],{"href":222,"rel":223},"https://www.nist.gov/itl/ai-risk-management-framework",[200],"NIST AI Risk Management Framework"," says the same thing in public-sector language: Govern, Map, Measure, Manage. Mapping systems and measuring incidents still fail if the product people click can write to Salesforce without a named signer.",[227,228,230],"h2",{"id":229},"words-youll-hear","Words you’ll hear",[232,233,234,240,256,267,273,279,295,305,317],"ul",{},[192,235,236,239],{},[178,237,238],{},"Live business system."," CRM, ERP, HR, billing — the tools that hold official numbers and customer records. At work, this is where a fluent sentence becomes a fact other teams will inherit.",[192,241,242,245,246,250,251,255],{},[178,243,244],{},"Write / write-back."," The AI is allowed to ",[247,248,249],"em",{},"change"," that system, not only draft a suggestion. See ",[196,252,254],{"href":253},"what-is-write-back-governance","What is write-back governance",". At work, a next-step note and an Amount field are not the same risk class.",[192,257,258,261,262,266],{},[178,259,260],{},"Human-in-the-loop."," A person must approve before the job can finish. See ",[196,263,265],{"href":264},"what-is-human-in-the-loop-ai","What is human-in-the-loop AI",". At work, the gate shows the payload in the language of the live system, not a wall of prompt text.",[192,268,269,272],{},[178,270,271],{},"Named signer."," The identity that authorised the change. At work, “someone in the channel clicked yes” is not a signer.",[192,274,275,278],{},[178,276,277],{},"Fail-closed."," Missing approval means nothing happens. Fail-open means the change goes through unless someone happens to stop it.",[192,280,281,284,285,288,289,294],{},[178,282,283],{},"DPIA."," A data-protection impact assessment — thinking through purpose, risk, and personal data ",[247,286,287],{},"before"," you turn a tool loose. ",[196,290,293],{"href":291,"rel":292},"https://ico.org.uk/for-organisations/uk-gdpr-guidance-and-resources/artificial-intelligence/",[200],"UK ICO guidance on AI and data protection"," still wants a lawful basis and purpose when the “user” is an AI.",[192,296,297,300,301,202],{},[178,298,299],{},"Shadow AI."," Personal ChatGPT for work because the official path is missing. See ",[196,302,304],{"href":303},"what-is-shadow-ai","What is shadow AI",[192,306,307,310,311,316],{},[178,308,309],{},"Inventory."," A list of where AI actually runs. The ",[196,312,315],{"href":313,"rel":314},"https://www.justice.gov/media/1373026/dl",[200],"US plan described in OMB M-24-10"," puts a named owner and an inventory first, not a PDF.",[192,318,319,322,323,326,327,331],{},[178,320,321],{},"Least privilege."," Only the data and tools required for ",[247,324,325],{},"this"," job. A ",[196,328,330],{"href":329},"what-is-an-ai-workstream","workstream"," is how that instinct becomes a company object.",[174,333,334,335,338,339,341],{},"Model safety is adjacent and different. Safety is about what a model will say in the abstract. Enterprise governance is about what ",[247,336,337],{},"your"," people and tools may do with ",[247,340,337],{}," systems and data. You can have a carefully aligned model and still have ungoverned CRM writes.",[227,343,345],{"id":344},"why-you-should-care","Why you should care",[174,347,348,349,354],{},"Without working rules, AI becomes a side effect. A field moves. A journal posts. A customer is told a policy the company does not hold. Nobody can say who allowed it. In February 2024, a British Columbia tribunal held Air Canada responsible for a chatbot that invented a bereavement-fare policy. ",[196,350,353],{"href":351,"rel":352},"https://www.cbc.ca/news/canada/british-columbia/air-canada-chatbot-lawsuit-1.7116416",[200],"CBC reported"," that the airline’s argument — the chatbot is a separate legal entity — failed. A customer-facing commitment without a working gate is still the company’s commitment.",[174,356,357],{},"It affects you if you:",[232,359,360,366,372,384],{},[192,361,362,365],{},[178,363,364],{},"Own a number."," Forecasts and close packs inherit whatever changed.",[192,367,368,371],{},[178,369,370],{},"Own a customer relationship."," Model output that becomes a commitment is still the company’s commitment.",[192,373,374,377,378,383],{},[178,375,376],{},"Own risk or legal."," Privacy law does not pause for a chatbot. ",[196,379,382],{"href":380,"rel":381},"https://eur-lex.europa.eu/eli/reg/2016/679/oj",[200],"GDPR"," still applies to purpose, minimisation, and erasure.",[192,385,386,389],{},[178,387,388],{},"Are asked “who is in charge of AI here?”"," An inventory and a named owner beat a principles slide.",[174,391,392],{},"Good governance in practice is four working rules:",[232,394,395,401,407,413],{},[192,396,397,400],{},[178,398,399],{},"People and rights."," Humans and AI tools are both actors. Roles decide what they may start, see, and sign.",[192,402,403,406],{},[178,404,405],{},"Data at question time."," Purpose and minimisation still apply when an AI is the one looking.",[192,408,409,412],{},[178,410,411],{},"Action rights."," Read-only is a control. Unrestricted tools are an incident waiting for a bad prompt.",[192,414,415,418,419,202],{},[178,416,417],{},"Evidence and spend."," Chat scrollback is not a management system. Uncapped spend is a budget failure and often a security failure. See ",[196,420,422],{"href":421},"what-is-ai-token-economics","What is AI token economics",[174,424,425,426,429],{},"Blocking consumer ChatGPT at the office network, while people use personal phones, is not governance. It is a ",[196,427,428],{"href":303},"shadow AI"," problem with extra steps.",[431,432,434],"h3",{"id":433},"what-changes-by-role","What changes by role",[174,436,437,440],{},[178,438,439],{},"Finance."," Governance is whether an AI-proposed journal can post, against which checklist, with which signer, and whether the spend of the run was capped. “Unlimited AI” is not a control. Surprise inference bills are a governance failure that looks like a cloud invoice.",[174,442,443,446,447,452],{},[178,444,445],{},"Legal."," Lawful basis, purpose limitation, customer-facing language, and reconstructable authorisation. Legal also has to separate the OECD-style public commitment from the runtime. A principles page does not implement Article-style oversight. For higher-risk systems, ",[196,448,451],{"href":449,"rel":450},"https://eur-lex.europa.eu/eli/reg/2024/1689/oj",[200],"EU AI law"," Article 14 talks about effective oversight: people must be able to interpret outputs and interrupt the system. A footer that says “generated by AI” is not that.",[174,454,455,458],{},[178,456,457],{},"Operations."," Isolation of jobs, connector scope, and a place to put a paused run. Ops already runs change control. Governance is change control that includes a model as a proposer.",[174,460,461,464],{},[178,462,463],{},"Go-to-market."," The difference between a draft email and a sent commitment; between a suggested next step and a changed Amount. GTM feels friction first. The honest metric is time-to-approved-write, not time-to-first-answer.",[174,466,467,470,471,476],{},[178,468,469],{},"Security."," Identity of the connected user, read versus write, prompt injection as a path to a tool call, and the new store created by logs and indexes. The ",[196,472,475],{"href":473,"rel":474},"https://genai.owasp.org/llm-top-10/",[200],"OWASP Top 10 for LLM applications"," treats retrieval and tool use as a security surface, not only a quality issue. Network DLP helps with paste-out. It does not quote a CRM change.",[431,478,480],{"id":479},"what-people-get-wrong","What people get wrong",[174,482,483,486],{},[178,484,485],{},"Governance as a committee."," Useful for risk registers. Useless if the product can still write.",[174,488,489,492],{},[178,490,491],{},"Governance as model safety."," Refusals on public-web questions do not bind Salesforce.",[174,494,495,498],{},[178,496,497],{},"Governance as a secure web gateway."," Necessary for some paste-out paths. Insufficient for writes, approvals, and causal history.",[174,500,501,504,505,202],{},[178,502,503],{},"Governance as blocking."," Blocks without a sanctioned path train people onto phones. See ",[196,506,304],{"href":303},[174,508,509,512],{},[178,510,511],{},"Theatre."," A checkbox, a prompt that says “ask first,” or an admin toggle the model can ignore.",[174,514,515,516,520],{},"Good looks like: connectors default to read-only; writes are quoted; a named signer cannot be waived by the model; evidence lives on a ",[196,517,519],{"href":518},"what-is-a-lifecycle-graph","lifecycle graph","; spend has a ceiling; scope follows the job. Failure looks like a PDF, a blocked URL, and a personal API key in a wiki.",[174,522,523,524,527,528,531,532,534,535,539],{},"Adjacent concepts: ",[196,525,526],{"href":253},"write-back governance"," is the write subset. ",[196,529,530],{"href":264},"Human-in-the-loop"," is the gate. ",[196,533,31],{"href":329}," are the isolation unit. An ",[196,536,538],{"href":537},"what-is-an-enterprise-ai-operating-system","enterprise AI operating system"," is the product shape that makes those rules the default path.",[227,541,543],{"id":542},"how-this-shows-up-in-nimbus","How this shows up in Nimbus",[174,545,546],{},"Nimbus treats governance as how work is released, not as a sidecar policy engine.",[174,548,549,550,553],{},"Connectors — secure links to live systems — default to ",[178,551,552],{},"read-only",". When a change is proposed, the product shows the intended action and waits. A named person must sign. The model cannot waive the gate. Missing approval is fail-closed: nothing happens.",[174,555,556,557,559,560,562,563,567],{},"Scope is the ",[196,558,330],{"href":329},": one job, with the playbooks, systems, teams, and budget that belong to that job. Evidence is the ",[196,561,23],{"href":518},". The ",[196,564,566],{"href":565},"what-is-a-company-wiki-for-ai-agents","company wiki"," is the asserted policy the run must cite. Model routing does not bypass the gate.",[174,569,570,571,573,574,202],{},"See ",[196,572,39],{"href":40},". For scoring vendors: ",[196,575,577],{"href":576},"how-to-evaluate-ai-governance-platforms","How to evaluate AI governance platforms",[227,579,581],{"id":580},"questions-people-actually-ask","Questions people actually ask",[431,583,585],{"id":584},"is-ai-governance-the-same-as-making-the-model-safe","Is AI governance the same as making the model “safe”?",[174,587,588,589,338,591,593],{},"No. Model safety is about what the model will say in the abstract. Enterprise governance is about what ",[247,590,337],{},[247,592,337],{}," systems and data.",[431,595,597],{"id":596},"can-we-rely-on-the-secure-web-gateway","Can we rely on the secure web gateway?",[174,599,600],{},"Network controls help with paste-out. They do not quote a CRM change, bind an approver, or store a causal history. Use both.",[431,602,604],{"id":603},"must-a-person-always-approve","Must a person always approve?",[174,606,607,608,611],{},"For many operational writes, yes. For read-only analysis, maybe not. The mistake is calling a system “human-approved” because a human ",[247,609,610],{},"could"," look, while changes proceed on model initiative.",[431,613,615],{"id":614},"do-the-oecd-ai-principles-require-a-specific-product","Do the OECD AI Principles require a specific product?",[174,617,618],{},"No. They are a public commitment. A product can make evidence cheaper to produce. The commitment does not implement a gate.",[431,620,622],{"id":621},"is-a-dpia-enough-to-go-live","Is a DPIA enough to go live?",[174,624,625],{},"It is necessary thinking, not a runtime. You still need identity, scope, fail-closed writes, and a record. The DPIA should describe those controls, not replace them.",[431,627,629],{"id":628},"does-blocking-chatgpt-count-as-governance","Does blocking ChatGPT count as governance?",[174,631,632,633,636],{},"It is a network control. Without a sanctioned path that can see the right files, people use personal phones. Blocking can tighten ",[247,634,635],{},"after"," substitution exists.",[431,638,640],{"id":639},"how-is-this-different-from-it-change-management","How is this different from IT change management?",[174,642,643],{},"It is the same instinct — who may change production, with what evidence — applied to a proposer that speaks English. Existing CAB processes rarely see model-initiated payloads unless the product emits them.",[431,645,647],{"id":646},"who-should-be-the-named-owner-of-ai","Who should be the named owner of AI?",[174,649,650],{},"Someone who can inventory systems and stop a write path, not a volunteer “champion” with no authority over CRM. Federal-style guidance starts with inventory and ownership for a reason.",[431,652,654],{"id":653},"can-we-govern-only-customer-facing-chatbots-and-ignore-internal-copilots","Can we govern only customer-facing chatbots and ignore internal copilots?",[174,656,657],{},"Internal tools still process personal data and still write to live systems. Air Canada was customer-facing. Ungoverned CRM hygiene is an internal path to the same class of invented fact.",[431,659,661],{"id":660},"do-we-need-the-eu-ai-act-if-we-are-not-a-high-risk-provider","Do we need the EU AI Act if we are not a high-risk provider?",[174,663,664],{},"You may still have GDPR duties, sector rules, and customer contracts. Oversight and records are useful even when a specific Act title does not apply. Do not claim “Act compliant” because you have a button.",[431,666,668],{"id":667},"where-does-spend-fit","Where does spend fit?",[174,670,671,672,202],{},"Uncapped inference is a control failure. Quotes, ceilings, and attribution by job are governance of a scarce, abusable resource. See ",[196,673,422],{"href":421},[431,675,677],{"id":676},"is-an-acceptable-use-policy-still-worth-writing","Is an acceptable-use policy still worth writing?",[174,679,680],{},"Yes, as communication. No, as enforcement. Write the PDF. Then put the same rules in the product people actually use.",[227,682,684],{"id":683},"related-reading","Related reading",[174,686,687,689,690,692,693,202],{},[196,688,254],{"href":253},", ",[196,691,304],{"href":303},", and ",[196,694,695],{"href":537},"What is an enterprise AI operating system",[227,697,699],{"id":698},"sources","Sources",[232,701,702,708,713,719,724,730,735,741],{},[192,703,704],{},[196,705,707],{"href":216,"rel":706},[200],"Gartner, AI governance and TRiSM",[192,709,710],{},[196,711,201],{"href":198,"rel":712},[200],[192,714,715],{},[196,716,718],{"href":291,"rel":717},[200],"ICO, AI and data protection",[192,720,721],{},[196,722,224],{"href":222,"rel":723},[200],[192,725,726],{},[196,727,729],{"href":351,"rel":728},[200],"CBC, Air Canada chatbot lawsuit",[192,731,732],{},[196,733,382],{"href":380,"rel":734},[200],[192,736,737],{},[196,738,740],{"href":449,"rel":739},[200],"EU AI Act (Regulation 2024/1689)",[192,742,743],{},[196,744,475],{"href":473,"rel":745},[200],{"title":155,"searchDepth":156,"depth":156,"links":747},[748,749,754,755,769,770],{"id":229,"depth":156,"text":230},{"id":344,"depth":156,"text":345,"children":750},[751,753],{"id":433,"depth":752,"text":434},3,{"id":479,"depth":752,"text":480},{"id":542,"depth":156,"text":543},{"id":580,"depth":156,"text":581,"children":756},[757,758,759,760,761,762,763,764,765,766,767,768],{"id":584,"depth":752,"text":585},{"id":596,"depth":752,"text":597},{"id":603,"depth":752,"text":604},{"id":614,"depth":752,"text":615},{"id":621,"depth":752,"text":622},{"id":628,"depth":752,"text":629},{"id":639,"depth":752,"text":640},{"id":646,"depth":752,"text":647},{"id":653,"depth":752,"text":654},{"id":660,"depth":752,"text":661},{"id":667,"depth":752,"text":668},{"id":676,"depth":752,"text":677},{"id":683,"depth":156,"text":684},{"id":698,"depth":156,"text":699},"2026-08-17","AI governance is the working rules for who may use which AI, on which data, and whether it may change a live business system — plus a record of what happened.","/blog/what-is-ai-governance",{"title":165,"description":772},"explainer","blog/what-is-ai-governance",[775,778,779,780],"governance","compliance","audit","8Gc6YA0kBilCFVsDbCgM_sj0gjR1GSQazeF0iy3Fot8",{"hero":783,"id":785,"title":786,"archived":149,"authors":150,"badge":150,"body":787,"date":150,"department":150,"description":791,"extension":158,"eyebrow":792,"faqHeader":150,"faqs":150,"footerBand":793,"headline":150,"image":150,"industry":150,"jobType":150,"listed":149,"location":150,"navigation":115,"openRoles":150,"pageLayout":150,"path":60,"relatedHeading":799,"seo":800,"series":150,"sitemap":115,"status":150,"stem":801,"subhead":150,"tags":150,"video":150,"whyJoin":150,"workplaceType":150,"__hash__":802},{"filename":784},"u2221455217_Flat_design_of_a_futuristic_minimalist_landscape__5d589295-cdea-4ea9-a262-be766881accf_1.png","content/blog/index.md","Exploring the future of intelligence.",{"type":152,"value":788,"toc":789},[],{"title":155,"searchDepth":156,"depth":156,"links":790},[],"Deep dives into pre-cognitive intelligence, sentient enterprises, and the evolving landscape of AI-driven business transformation.","Latest Research",{"headline":794,"description":795,"primaryLabel":796,"primaryTo":797,"secondaryLabel":798,"secondaryTo":12},"Stay at the frontier.","Subscribe for product updates and new insights.","Subscribe","/newsletter","Explore the platform","More research",{"title":786,"description":791},"blog/index","eK1RCXdDW8nfLSyKRXGB1mJm9FAmhAO6GWXwNKOMNVE",[804,1264],{"id":805,"title":806,"archived":149,"authors":807,"badge":809,"body":810,"date":771,"department":150,"description":1255,"extension":158,"eyebrow":150,"faqHeader":150,"faqs":150,"footerBand":150,"headline":150,"image":150,"industry":150,"jobType":150,"listed":149,"location":150,"navigation":115,"openRoles":150,"pageLayout":150,"path":1256,"relatedHeading":150,"seo":1257,"series":775,"sitemap":115,"status":150,"stem":1258,"subhead":150,"tags":1259,"video":150,"whyJoin":150,"workplaceType":150,"__hash__":1263},"content/blog/what-is-ai-token-economics.md","What is AI Token Economics",[808],{"name":168,"to":120},{"label":170},{"type":152,"value":811,"toc":1231},[812,831,837,840,847,849,935,946,948,950,970,973,996,999,1001,1010,1015,1020,1025,1033,1035,1041,1051,1057,1065,1077,1086,1088,1091,1097,1106,1108,1112,1115,1119,1122,1126,1129,1133,1138,1142,1145,1149,1152,1156,1161,1165,1168,1172,1175,1179,1184,1188,1193,1197,1200,1202,1209,1211],[174,813,814,815,818,819,824,825,830],{},"A ",[178,816,817],{},"token"," is a chunk of text the model reads or writes. You pay per chunk. Different models cost different amounts. Input, output, and sometimes tools all meter differently. ",[196,820,823],{"href":821,"rel":822},"https://openai.com/api/pricing/",[200],"OpenAI"," and ",[196,826,829],{"href":827,"rel":828},"https://www.anthropic.com/pricing",[200],"Anthropic"," publish those ladders. Finance still cannot run the business on “12 million tokens of vendor A’s flagship.”",[174,832,833,836],{},[178,834,835],{},"AI token economics"," is treating that usage like a real budget: measuring, allocating, controlling, and attributing spend so operators can quote before a run, cap during it, and attribute after it — instead of a slide that says “unlimited AI.”",[174,838,839],{},"Without it, organisations either freeze (no production AI) or send every small task to the most expensive model until the bill becomes a board slide.",[174,841,842,843,846],{},"The unit problem is the same one cloud had in its first decade: a metered resource sold with a headcount story. Seat licences predict people. Inference predicts work. When those two are collapsed into “unlimited,” the next chunk ",[247,844,845],{},"feels"," free, so people pick the flagship every time. The ladder did not disappear. It hid.",[227,848,230],{"id":229},[232,850,851,857,863,872,881,887,893,902,908,919,925],{},[192,852,853,856],{},[178,854,855],{},"Token."," A piece of text the model processes. Not a business unit. At work, a long wiki dump and a short field extract are wildly different token counts for the same “question.”",[192,858,859,862],{},[178,860,861],{},"Seat licence."," Predictable cost per person. Often marketed as “unlimited.” The underlying work is still metered.",[192,864,865,868,869,871],{},[178,866,867],{},"Pass-through API bill."," Each team has keys. Simple. Invites key sprawl and ",[196,870,428],{"href":303}," on personal keys. At work, the invoice lands in engineering while go-to-market did the looping.",[192,873,874,877,878,880],{},[178,875,876],{},"Quote."," A number ",[247,879,287],{}," they run. At work, this is what makes a brief a decision rather than a surprise.",[192,882,883,886],{},[178,884,885],{},"Cap / ceiling."," A hard stop. The loop cannot spend past it. At work, weekend agent loops die here instead of in next month’s cloud bill.",[192,888,889,892],{},[178,890,891],{},"Pool."," Organisation-level allowance. At work, one department should not be able to burn the company pool on a vanity run.",[192,894,895,898,899,901],{},[178,896,897],{},"Attribution."," Chargeback by job, not “the AI bill.” At work, finance can ask which ",[196,900,330],{"href":329}," consumed the units.",[192,903,904,907],{},[178,905,906],{},"NTU (Nimbus Token Unit)."," Nimbus’s normalised work credit for completed AI activity — analysis, tools, runs, writes — sitting above raw provider tokens. Everyday questions can be included; heavier work consumes pool credits. Finance gets one tape measure across vendors and steps.",[192,909,910,913,914,918],{},[178,911,912],{},"Model routing."," Cheaper model for simple steps, stronger only when needed. See ",[196,915,917],{"href":916},"what-is-model-routing","What is model routing",". At work, classify-this-ticket should not pay flagship rates.",[192,920,921,924],{},[178,922,923],{},"Context window."," How much text the model can see at once. Dumping the whole Drive into context is an economic choice, not a quality strategy.",[192,926,927,930,931,202],{},[178,928,929],{},"Stop condition."," Budget hit, empty result, human cancel. Agent loops can dominate the bill without improving the artefact. See ",[196,932,934],{"href":933},"what-is-an-agentic-workflow","What is an agentic workflow",[174,936,937,938,941,942,945],{},"The point is ",[178,939,940],{},"value per unit",", not minimum units regardless of outcome. Caching, wiki citations, and memory should make the ",[247,943,944],{},"same"," outcome cheaper over time. If unit cost of an approved update never falls, you are re-deriving folklore every run.",[227,947,345],{"id":344},[174,949,357],{},[232,951,952,958,964],{},[192,953,954,957],{},[178,955,956],{},"Own the budget."," Surprise invoices arrive after agents looped all weekend.",[192,959,960,963],{},[178,961,962],{},"Run the work."," You should see a number before you commit, not a lecture after.",[192,965,966,969],{},[178,967,968],{},"Are tempted to shame people for using AI."," Shame drives personal keys. Cap the official path so it is safe to use.",[174,971,972],{},"Practical rhythm:",[232,974,975,984,990],{},[192,976,977,980,981,983],{},[178,978,979],{},"Name the run."," Unnamed chats cannot be attributed. That is what a ",[196,982,330],{"href":329}," is for.",[192,985,986,989],{},[178,987,988],{},"Separate exploration from production."," Sandboxes can have tighter caps and cheaper default routes.",[192,991,992,995],{},[178,993,994],{},"Review unit cost of outcomes"," — approved updates per unit — not tokens in the abstract.",[174,997,998],{},"Anti-pattern: a single corporate API key in a wiki, no per-job cap, monthly surprise. That is an unmetered utility.",[431,1000,434],{"id":433},[174,1002,1003,1005,1006,1009],{},[178,1004,439],{}," You need a quote, a ceiling, and a chargeback dimension that matches how the business already thinks — by job, department, or cost centre — not by vendor token type. Multi-vendor ladders are incomparable until you normalise. NTU is that normalisation in Nimbus. Finance should also see ",[247,1007,1008],{},"stops",": a cap that fired is a successful control, not a failed project.",[174,1011,1012,1014],{},[178,1013,445],{}," Spend logs are not only money. They are a map of which data classes went to which provider. Uncapped personal keys are a processing-agreement gap. Legal will also ask whether you can stop a run, not only whether you can pay for it.",[174,1016,1017,1019],{},[178,1018,457],{}," Caps are operational stops, like a queue limit. Ops needs to know whether a paused run is waiting on a person or waiting on budget. Mixing those two in one “it failed” status is how you get the wrong pager.",[174,1021,1022,1024],{},[178,1023,463],{}," GTM feels the quality-versus-cost trade first. A compact model that extracts fields is usually enough. A flagship model that argues a clause may be worth it. Without routing and quotes, GTM either hoards “the best model” or gets blamed for the bill. Neither produces better pipeline hygiene.",[174,1026,1027,1029,1030,1032],{},[178,1028,469],{}," API keys are credentials. Personal keys in browser plugins are ",[196,1031,428],{"href":303},". A pooled official path with per-workstream ceilings reduces key sprawl. Spend spikes can also be an anomaly signal — a loop that never stops is sometimes a bug, sometimes a prompt-injection success.",[431,1034,480],{"id":479},[174,1036,1037,1040],{},[178,1038,1039],{},"“Unlimited” as a strategy."," Seats hide the ladder. They do not delete it. Heavy agentic work will still surface as a true-up, a throttle, or a degraded model.",[174,1042,1043,1046,1047,1050],{},[178,1044,1045],{},"Punishing usage."," Chargeback without a sanctioned path recreates personal keys. Celebrate lower units ",[247,1048,1049],{},"per artefact"," as playbooks and memory compound.",[174,1052,1053,1056],{},[178,1054,1055],{},"Tokens as the KPI."," Tokens measure consumption. Outcomes measure value. A cheap run that produces a rejected write is still waste. A dearer run that produces one approved journal may be fine.",[174,1058,1059,1062,1063,202],{},[178,1060,1061],{},"One model for everything."," That is a routing failure dressed as quality culture. See ",[196,1064,917],{"href":916},[174,1066,1067,1070,1071,1076],{},[178,1068,1069],{},"No stop on loops."," ",[196,1072,1075],{"href":1073,"rel":1074},"https://www.anthropic.com/engineering/building-effective-agents",[200],"Anthropic’s note on building effective agents"," treats workflows with stop conditions as the grown-up shape. Economics is one of those stops.",[174,1078,1079,1080,824,1083,1085],{},"Good looks like: named jobs, quotes before commit, hard ceilings, routing policy, attribution, and falling unit cost as the ",[196,1081,1082],{"href":565},"wiki",[196,1084,519],{"href":518}," reduce re-derivation. Failure looks like a shared key, a flagship default, and a board slide titled “AI spend.”",[227,1087,543],{"id":542},[174,1089,1090],{},"Workstreams show quotes and ceilings before runs. Orgs draw from a pooled NTU allowance. Routing is a policy, not a dropdown labelled “best.” Memory and wiki reduce re-derivation, which is how unit cost of an outcome should fall over time.",[174,1092,1093,1094,1096],{},"Everyday questions can sit inside the allowance; heavier analysis, tools, and writes consume pool credits. The ",[196,1095,23],{"href":518}," can record spend as part of the chain, so “the run stopped because the ceiling was hit” is a causal fact.",[174,1098,570,1099,824,1101,1103,1104,202],{},[196,1100,44],{"href":45},[196,1102,917],{"href":916},". Product context: ",[196,1105,31],{"href":32},[227,1107,581],{"id":580},[431,1109,1111],{"id":1110},"why-cant-we-just-pay-seats-and-call-it-unlimited","Why can’t we just pay seats and call it unlimited?",[174,1113,1114],{},"Seats predict headcount. Production AI spend is inference, tools, and writes. “Unlimited” hides the ladder; it does not delete it.",[431,1116,1118],{"id":1117},"what-should-finance-actually-see","What should finance actually see?",[174,1120,1121],{},"A quote before commit, a cap during the run, and attribution by job afterwards — in one unit they can compare across vendors and steps.",[431,1123,1125],{"id":1124},"wont-cheaper-models-get-worse-answers","Won’t cheaper models get worse answers?",[174,1127,1128],{},"For extract and classify, often no. For hard judgment, often yes. That is a routing policy, not a religion. Measure reject rates on the job, not vibes.",[431,1130,1132],{"id":1131},"do-we-punish-teams-for-using-ai","Do we punish teams for using AI?",[174,1134,1135,1136,1050],{},"No. Punishing usage revives shadow AI. Celebrate lower units ",[247,1137,1049],{},[431,1139,1141],{"id":1140},"what-is-an-ntu-in-plain-language","What is an NTU in plain language?",[174,1143,1144],{},"A normalised work credit above raw provider tokens, so a finance partner is not asked to compare “vendor A input tokens” with “vendor B output tokens” plus tool calls. In Nimbus, completed activity — analysis, tools, runs, writes — is what consumes the unit.",[431,1146,1148],{"id":1147},"should-every-chat-be-billed-to-a-cost-centre","Should every chat be billed to a cost centre?",[174,1150,1151],{},"Named production jobs, yes. Tiny sanctioned copilots for personal drafting can live on a lighter path. The failure is mixing them so neither can be capped.",[431,1153,1155],{"id":1154},"how-do-agent-loops-blow-the-budget","How do agent loops blow the budget?",[174,1157,1158,1159,202],{},"They call tools, re-read context, and retry without a finish line. Without a ceiling and a stop condition, “being thorough” is an unbounded loop. See ",[196,1160,934],{"href":933},[431,1162,1164],{"id":1163},"is-caching-the-same-as-token-economics","Is caching the same as token economics?",[174,1166,1167],{},"Caching is a tactic. Economics is the management system: quote, cap, attribute, route. Caching without attribution still leaves you unable to explain the bill.",[431,1169,1171],{"id":1170},"do-we-need-a-data-warehouse-to-do-this","Do we need a data warehouse to do this?",[174,1173,1174],{},"You need events at run time. A warehouse can hold copies for reporting. It cannot quote a run that has not emitted a number yet.",[431,1176,1178],{"id":1177},"how-does-this-relate-to-write-back","How does this relate to write-back?",[174,1180,1181,1182,202],{},"Writes are usually a small number of tokens and a large operational risk. Do not use spend as a substitute for a named signer. Do use spend as a stop so a looping agent cannot keep proposing writes all weekend. See ",[196,1183,254],{"href":253},[431,1185,1187],{"id":1186},"can-we-lock-one-vendor-to-simplify-pricing","Can we lock one vendor to simplify pricing?",[174,1189,1190,1191,202],{},"You can. You will pay for it in price, outages, and lock-in. A normalised unit plus routing is how finance keeps a second tape measure. See ",[196,1192,917],{"href":916},[431,1194,1196],{"id":1195},"why-not-just-set-a-monthly-company-cap","Why not just set a monthly company cap?",[174,1198,1199],{},"A company cap without per-job attribution is a shared kitchen. The loudest workflow starves the others, and nobody can say which job did it.",[227,1201,684],{"id":683},[174,1203,1204,824,1206,202],{},[196,1205,917],{"href":916},[196,1207,1208],{"href":329},"What is an AI workstream",[227,1210,699],{"id":698},[232,1212,1213,1219,1225],{},[192,1214,1215],{},[196,1216,1218],{"href":821,"rel":1217},[200],"OpenAI API pricing",[192,1220,1221],{},[196,1222,1224],{"href":827,"rel":1223},[200],"Anthropic pricing",[192,1226,1227],{},[196,1228,1230],{"href":1073,"rel":1229},[200],"Anthropic, Building effective agents",{"title":155,"searchDepth":156,"depth":156,"links":1232},[1233,1234,1238,1239,1253,1254],{"id":229,"depth":156,"text":230},{"id":344,"depth":156,"text":345,"children":1235},[1236,1237],{"id":433,"depth":752,"text":434},{"id":479,"depth":752,"text":480},{"id":542,"depth":156,"text":543},{"id":580,"depth":156,"text":581,"children":1240},[1241,1242,1243,1244,1245,1246,1247,1248,1249,1250,1251,1252],{"id":1110,"depth":752,"text":1111},{"id":1117,"depth":752,"text":1118},{"id":1124,"depth":752,"text":1125},{"id":1131,"depth":752,"text":1132},{"id":1140,"depth":752,"text":1141},{"id":1147,"depth":752,"text":1148},{"id":1154,"depth":752,"text":1155},{"id":1163,"depth":752,"text":1164},{"id":1170,"depth":752,"text":1171},{"id":1177,"depth":752,"text":1178},{"id":1186,"depth":752,"text":1187},{"id":1195,"depth":752,"text":1196},{"id":683,"depth":156,"text":684},{"id":698,"depth":156,"text":699},"AI token economics is treating AI usage like a real budget: you pay per chunk of text the model reads and writes, so finance can quote, cap, and attribute spend instead of hoping for “unlimited AI.”","/blog/what-is-ai-token-economics",{"title":806,"description":1255},"blog/what-is-ai-token-economics",[775,1260,1261,1262],"token-economics","ntu","model-routing","kgDCkz6CVtDSpJc699QfUH1nMlvaVezkE2PrGzu2Puc",{"id":1265,"title":1266,"archived":149,"authors":1267,"badge":1269,"body":1270,"date":771,"department":150,"description":1717,"extension":158,"eyebrow":150,"faqHeader":150,"faqs":150,"footerBand":150,"headline":150,"image":150,"industry":150,"jobType":150,"listed":149,"location":150,"navigation":115,"openRoles":150,"pageLayout":150,"path":1718,"relatedHeading":150,"seo":1719,"series":775,"sitemap":115,"status":150,"stem":1720,"subhead":150,"tags":1721,"video":150,"whyJoin":150,"workplaceType":150,"__hash__":1724},"content/blog/what-is-a-lifecycle-graph.md","What is a Lifecycle Graph",[1268],{"name":168,"to":120},{"label":170},{"type":152,"value":1271,"toc":1692},[1272,1279,1282,1285,1288,1290,1346,1367,1369,1372,1375,1377,1403,1418,1420,1427,1432,1437,1442,1447,1451,1454,1471,1474,1477,1484,1486,1493,1504,1511,1520,1523,1525,1531,1537,1544,1554,1556,1560,1567,1571,1577,1581,1584,1588,1591,1595,1600,1604,1607,1611,1614,1618,1624,1628,1631,1635,1638,1642,1647,1651,1658,1660,1670,1672],[174,1273,1274,1275,1278],{},"When people ask what a lifecycle graph is, they are usually asking about ",[178,1276,1277],{},"causality",": why did this number, field, or decision change?",[174,1280,1281],{},"Causality is the difference between “two things happened around the same time” and “this caused that.” If pipeline coverage went up in the same month AI usage went up, that is a coincidence until you can show the actual steps: what was asked, what was used, who approved it, and what the live system did.",[174,1283,1284],{},"A lifecycle graph is the company’s record of those steps. It is not a chat history. Chat history shows that someone talked to a model. A graph shows the chain from the question to the outcome, so the next person — or an auditor — can follow it.",[174,1286,1287],{},"This is an operations problem that existed before generative AI. ERP journals already needed authorisation trails. CRM already had field history. What changed is that a new kind of actor can now propose, and sometimes execute, those changes in fluent language. If the “why” lives only in a personal chat, the company has a causality gap the moment that person leaves, the vendor rotates logs, or the model version rolls.",[227,1289,230],{"id":229},[232,1291,1292,1298,1304,1316,1322,1328,1334,1340],{},[192,1293,1294,1297],{},[178,1295,1296],{},"Causality."," Being able to say what caused what, with evidence. At work, this is “this next-step field changed because this brief ran, cited this playbook version, and this person signed this payload.”",[192,1299,1300,1303],{},[178,1301,1302],{},"Correlation."," Two things moving together. Not the same as cause. At work, AI usage and pipeline moving in the same quarter is a slide, not an explanation.",[192,1305,1306,1309,1310,1315],{},[178,1307,1308],{},"Provenance."," The trail of who, what, and when behind a piece of data. ",[196,1311,1314],{"href":1312,"rel":1313},"https://www.w3.org/TR/prov-overview/",[200],"W3C PROV"," is the open standard for that idea: entities, activities, and agents, linked so you can reconstruct derivation. A lifecycle graph is that instinct applied to AI-mediated work, not a claim that you have implemented the full W3C stack.",[192,1317,1318,1321],{},[178,1319,1320],{},"System of record."," The official live system that holds the fact — Salesforce for an opportunity, NetSuite for a journal. The graph should point at that record, not become a second copy of it. At work, pointing is how you avoid a second CRM that nobody can delete.",[192,1323,1324,1327],{},[178,1325,1326],{},"Audit trail."," A log that something happened. Useful, but thin if it cannot join the question, the policy, the signer, and the change. Provider API logs are an audit trail of calls. They are not a story of the job.",[192,1329,1330,1333],{},[178,1331,1332],{},"Lineage."," Which sources fed which proposal. At work, “which wiki version and which CRM records were in scope when this quote was generated?”",[192,1335,1336,1339],{},[178,1337,1338],{},"Retention."," How long a class of node is kept. A journal that feeds the books may need years. A draft may need weeks.",[192,1341,1342,1345],{},[178,1343,1344],{},"Scope."," Which job’s chain you are allowed to see. At work, a go-to-market question should not surface People Ops briefs.",[174,1347,1348,1349,1352,1353,1357,1358,1361,1362,1366],{},"Keep this graph apart from two neighbours. A ",[178,1350,1351],{},"business knowledge graph"," models customers, products, and sites. ",[196,1354,1356],{"href":1355},"what-is-enterprise-rag","Enterprise RAG"," retrieves documents that ",[247,1359,1360],{},"exist",". ",[196,1363,1365],{"href":1364},"what-is-institutional-memory-in-enterprise-ai","Institutional memory"," is the broader goal — what the company still knows after people leave. The lifecycle graph is the decision-memory layer of that goal: the chain of AI-mediated work.",[227,1368,345],{"id":344},[174,1370,1371],{},"AI makes it easy to change company systems without leaving a story. Someone asks a model to tidy CRM notes. A field moves. Next quarter, finance or legal asks why. The person who asked has left. The “why” lived in a personal chat. The CRM only shows the new value.",[174,1373,1374],{},"That is a causality problem. You cannot manage what you cannot reconstruct.",[174,1376,357],{},[232,1378,1379,1385,1391,1397],{},[192,1380,1381,1384],{},[178,1382,1383],{},"Sign off on numbers."," Forecasts, journals, and board packs inherit whatever AI changed last month.",[192,1386,1387,1390],{},[178,1388,1389],{},"Inherit someone else’s work."," You need the exception, not a rumour that “we always do 18% for strategic accounts.”",[192,1392,1393,1396],{},[178,1394,1395],{},"Answer auditors or regulators."," They will not accept “the chatbot did it.”",[192,1398,1399,1402],{},[178,1400,1401],{},"Switch vendors or models."," Provider logs are the vendor’s artefact. They are not your company memory.",[174,1404,1405,1406,1409,1410,1414,1415],{},"This is not the same as proving that a discount ",[247,1407,1408],{},"caused"," a won deal. That is a statistics question. See ",[196,1411,1413],{"href":1412},"what-is-causal-ai-for-operations","What is causal AI for operations",". A lifecycle graph answers a more basic one: ",[178,1416,1417],{},"what did we actually do, and who caused it?",[431,1419,434],{"id":433},[174,1421,1422,1424,1425,202],{},[178,1423,439],{}," Close packs and forecasts inherit field history. If an AI-proposed journal posted, finance needs the brief, the playbook version, the named signer, and the ERP response — not a Slack screenshot. Spend also belongs on the chain: a run that stopped because a cap was hit is a causal fact, not a missing invoice. See ",[196,1426,422],{"href":421},[174,1428,1429,1431],{},[178,1430,445],{}," Exception language, customer commitments, and “who saw what” are discovery questions. A graph that points at the payload the signer saw is evidence. A chat export from a personal account is not. Legal also cares about retention and deletion: infinite chat fails a privacy review; typed retention with export and legal hold is how records programmes already work.",[174,1433,1434,1436],{},[178,1435,457],{}," Handoffs fail when the next shift cannot see why a run paused. Human wait is a node, not an interruption. Incident reviews need the same chain: which connector was read-only, which write was refused, which wiki page caused the flag.",[174,1438,1439,1441],{},[178,1440,463],{}," Pipeline hygiene and renewal exceptions are where “the bot updated it” becomes a forecast problem. GTM needs to see the quoted fields, not a summary that says “updated pricing.” They also need scope: one team’s competitive notes should not leak into another region’s chain.",[174,1443,1444,1446],{},[178,1445,469],{}," The graph is a sensitive store. It should not hold full transcripts with secrets by default, other teams’ out-of-scope work, or the model’s private scratch reasoning. Access control on the graph is as important as access control on the CRM. A query surface that ignores vaults recreates the god workspace.",[431,1448,1450],{"id":1449},"what-belongs-on-the-chain","What belongs on the chain",[174,1452,1453],{},"Keep the links that let a non-engineer reconstruct a change:",[232,1455,1456,1459,1462,1465,1468],{},[192,1457,1458],{},"the job and the question",[192,1460,1461],{},"the sources (which playbook version, which records, which files)",[192,1463,1464],{},"the people (the model is not an answer for “who”)",[192,1466,1467],{},"the proposed change, in the language of the live system — fields and values, not “updated pricing”",[192,1469,1470],{},"the decision, the timestamp, and whether the live system accepted it",[174,1472,1473],{},"Point at the CRM record and the policy page. Do not copy the whole company into the graph. Copies become a second official system, and a deletion problem.",[174,1475,1476],{},"Do not keep, by default: the model’s private scratch reasoning, full transcripts with secrets, or other teams’ work that was never in scope.",[174,1478,1479,1480,1483],{},"Retention should follow the type of record. A journal that feeds the books may need years. A draft may need weeks. “Keep everything forever because AI” fails a privacy review. ",[196,1481,293],{"href":291,"rel":1482},[200]," still wants purpose and minimisation when the “user” is an AI.",[431,1485,480],{"id":479},[174,1487,1488,1489,1492],{},"The first mistake is ",[178,1490,1491],{},"treating chat history as the record",". Chat is a user interface. It is not a join of brief, policy, signer, and system response.",[174,1494,1495,1496,1499,1500,1503],{},"The second is ",[178,1497,1498],{},"retrofitting",". Copying six months of ChatGPT and Slack into a warehouse is archaeology. You still need something that ",[247,1501,1502],{},"emits"," events at the moment of the brief, the quote, and the approval.",[174,1505,1506,1507,1510],{},"The third is ",[178,1508,1509],{},"a second CRM",". Duplicating every opportunity into the graph “for completeness” creates conflicting official numbers and an erasure nightmare.",[174,1512,1513,1514,1517,1518,202],{},"The fourth is ",[178,1515,1516],{},"confusing this with causal science",". A fluent model paragraph that says “because” is not identification. Neither is a dashboard of two rising lines. See ",[196,1519,1413],{"href":1412},[174,1521,1522],{},"Good looks like reconstructable interventions with pointers, named people, and typed retention. Failure looks like a vendor log, a personal thread, or an infinite lake of tokens.",[227,1524,543],{"id":542},[174,1526,1527,1528,1530],{},"Nimbus’s ",[178,1529,23],{}," is that chain as a product: briefs, playbook citations, connector reads, spend, approvals, and write results are linked as work proceeds.",[174,1532,1533,1536],{},[178,1534,1535],{},"Perception"," is how you ask it in ordinary language — “why did this opportunity change last month?” — instead of reconstructing Slack.",[174,1538,1539,1540,1543],{},"The graph records ",[178,1541,1542],{},"work",", not every token the company ever sent to a model. Scope follows the job, so a go-to-market question should not surface People Ops briefs. Connectors default to read-only; a read that did not write is itself a node worth knowing. Fail-closed writes mean a missing named signer is a recorded refusal, not a silent mutation.",[174,1545,1546,1547,824,1549,1551,1552,202],{},"Product: ",[196,1548,23],{"href":24},[196,1550,1535],{"href":36},". The job that produces the chain is a ",[196,1553,330],{"href":329},[227,1555,581],{"id":580},[431,1557,1559],{"id":1558},"is-this-just-a-knowledge-graph-of-the-business","Is this just a knowledge graph of the business?",[174,1561,1562,1563,1566],{},"No. A business knowledge graph models customers, products, and sites. A lifecycle graph models ",[178,1564,1565],{},"AI-mediated work"," — what was asked, who signed, what changed. They can link (the write points at an opportunity). They are not the same thing.",[431,1568,1570],{"id":1569},"cant-the-warehouse-be-the-record","Can’t the warehouse be the record?",[174,1572,1573,1574,1576],{},"You can copy events into a warehouse for reporting. You still need something that ",[247,1575,1502],{}," those events at the moment of the brief, the quote, and the approval. Retrofitting six months of ChatGPT and Slack is archaeology, not operations.",[431,1578,1580],{"id":1579},"how-is-this-different-from-the-model-providers-logs","How is this different from the model provider’s logs?",[174,1582,1583],{},"Provider logs show API calls. They do not know your job, your playbook version, your approver, or whether the write was rejected.",[431,1585,1587],{"id":1586},"how-long-should-we-keep-it","How long should we keep it?",[174,1589,1590],{},"Treat it like other control evidence. Align retention with the type of record, legal hold, and storage limits. The product must support export, deletion, and access control — not infinite chat.",[431,1592,1594],{"id":1593},"how-does-this-relate-to-a-person-having-to-approve","How does this relate to a person having to approve?",[174,1596,1597,1598,202],{},"A human gate only counts if you can later show who signed and what they saw. Without a graph, that gate is a popup that forgets. See ",[196,1599,265],{"href":264},[431,1601,1603],{"id":1602},"does-the-graph-replace-crm-field-history","Does the graph replace CRM field history?",[174,1605,1606],{},"No. Field history says the value changed. The graph says which job, which policy version, and which named signer caused the proposal. Keep both. Point; do not duplicate.",[431,1608,1610],{"id":1609},"what-if-the-models-explanation-disagrees-with-the-graph","What if the model’s explanation disagrees with the graph?",[174,1612,1613],{},"Trust the structure. Fluent “because” text is often written after the fact. The chain of brief, sources, quote, and signature is the operational cause.",[431,1615,1617],{"id":1616},"can-we-store-every-prompt-and-completion","Can we store every prompt and completion?",[174,1619,1620,1621,202],{},"You can. You usually should not. Completeness is reconstructability, not hoarding. Secrets in transcripts become a new breach class. See ",[196,1622,1623],{"href":1364},"What is institutional memory in enterprise AI",[431,1625,1627],{"id":1626},"how-do-permissions-work-on-the-graph","How do permissions work on the graph?",[174,1629,1630],{},"The same least-privilege instinct as the job. If you could not see the People Ops workstream, you should not query its chain in ordinary language either.",[431,1632,1634],{"id":1633},"is-a-screenshot-of-the-approval-enough","Is a screenshot of the approval enough?",[174,1636,1637],{},"For a one-off incident, maybe. For a control, no. Screenshots do not join, do not retain by type, and do not survive the laptop.",[431,1639,1641],{"id":1640},"where-does-spend-sit-on-the-chain","Where does spend sit on the chain?",[174,1643,1644,1645,202],{},"Quotes, caps, and stop-on-budget are causal events. “The run did not write because the ceiling was hit” is an answer finance can use. See ",[196,1646,422],{"href":421},[431,1648,1650],{"id":1649},"how-is-this-different-from-mlops-experiment-tracking","How is this different from MLOps experiment tracking?",[174,1652,1653,1654,1657],{},"MLOps tracks model training and deployment. A lifecycle graph tracks operational work that ",[247,1655,1656],{},"uses"," models. They stack. They do not substitute.",[227,1659,684],{"id":683},[174,1661,1662,1663,1551,1665,1667,1668,202],{},"If the goal is what the company still knows after people leave, read ",[196,1664,1623],{"href":1364},[196,1666,330],{"href":329},". For the science versus operations cut, ",[196,1669,1413],{"href":1412},[227,1671,699],{"id":698},[232,1673,1674,1680,1687],{},[192,1675,1676],{},[196,1677,1679],{"href":1312,"rel":1678},[200],"W3C PROV overview",[192,1681,1682],{},[196,1683,1686],{"href":1684,"rel":1685},"https://www.w3.org/TR/prov-dm/",[200],"W3C PROV data model",[192,1688,1689],{},[196,1690,718],{"href":291,"rel":1691},[200],{"title":155,"searchDepth":156,"depth":156,"links":1693},[1694,1695,1700,1701,1715,1716],{"id":229,"depth":156,"text":230},{"id":344,"depth":156,"text":345,"children":1696},[1697,1698,1699],{"id":433,"depth":752,"text":434},{"id":1449,"depth":752,"text":1450},{"id":479,"depth":752,"text":480},{"id":542,"depth":156,"text":543},{"id":580,"depth":156,"text":581,"children":1702},[1703,1704,1705,1706,1707,1708,1709,1710,1711,1712,1713,1714],{"id":1558,"depth":752,"text":1559},{"id":1569,"depth":752,"text":1570},{"id":1579,"depth":752,"text":1580},{"id":1586,"depth":752,"text":1587},{"id":1593,"depth":752,"text":1594},{"id":1602,"depth":752,"text":1603},{"id":1609,"depth":752,"text":1610},{"id":1616,"depth":752,"text":1617},{"id":1626,"depth":752,"text":1627},{"id":1633,"depth":752,"text":1634},{"id":1640,"depth":752,"text":1641},{"id":1649,"depth":752,"text":1650},{"id":683,"depth":156,"text":684},{"id":698,"depth":156,"text":699},"A lifecycle graph is how a company answers “why did this happen?” after AI is involved — the chain of cause and effect, not a chat log.","/blog/what-is-a-lifecycle-graph",{"title":1266,"description":1717},"blog/what-is-a-lifecycle-graph",[775,1722,1723,780],"lifecycle-graph","institutional-memory","h3JSROHGCfRknyBr-B5YQvfKgkgs4vl4qnJ6vzw_PHE",{"enabled":149,"message":1726,"linkLabel":78,"linkHref":79,"id":1727,"title":1728,"archived":149,"authors":150,"badge":150,"body":1729,"date":150,"department":150,"description":155,"extension":158,"eyebrow":150,"faqHeader":150,"faqs":150,"footerBand":150,"headline":150,"image":150,"industry":150,"jobType":150,"listed":149,"location":150,"navigation":115,"openRoles":150,"pageLayout":150,"path":1733,"relatedHeading":150,"seo":1734,"series":150,"sitemap":115,"status":150,"stem":1735,"subhead":150,"tags":150,"video":150,"whyJoin":150,"workplaceType":150,"__hash__":1736},"We're hiring! Join the team building the Sentient Enterprise.","content/shared/hiring.md","Hiring banner",{"type":152,"value":1730,"toc":1731},[],{"title":155,"searchDepth":156,"depth":156,"links":1732},[],"/shared/hiring",{"title":1728,"description":155},"shared/hiring","-6bioYD7lKYokGUVU3ff4hHTvB-sDyOMuCptKHnojfk",{"fold":1738,"id":1742,"title":1743,"archived":149,"authors":150,"badge":150,"body":1744,"date":150,"department":150,"description":155,"extension":158,"eyebrow":150,"faqHeader":150,"faqs":150,"footerBand":1748,"headline":150,"image":150,"industry":150,"jobType":150,"listed":149,"location":150,"navigation":115,"openRoles":150,"pageLayout":150,"path":1752,"relatedHeading":150,"seo":1753,"series":150,"sitemap":115,"status":150,"stem":1754,"subhead":150,"tags":150,"video":150,"whyJoin":150,"workplaceType":150,"__hash__":1755},{"headline":1739,"description":1740,"primaryLabel":8,"primaryTo":1741,"secondaryLabel":798,"secondaryTo":12},"Run frontier AI your business actually owns.","Governed agent swarms, 2,000+ integrations, and a knowledge graph that stays inside your walls. Free 7-day trial.","/checkout","content/shared/cta.md","Site CTAs",{"type":152,"value":1745,"toc":1746},[],{"title":155,"searchDepth":156,"depth":156,"links":1747},[],{"headline":1749,"description":1750,"primaryLabel":8,"primaryTo":1741,"secondaryLabel":1751,"secondaryTo":84},"See what governed AI looks like on your stack.","Connect your tools, run a workstream, and keep every decision on your ledger - free for 7 days.","Talk to our team","/shared/cta",{"title":1743,"description":155},"shared/cta","YHK6Fb8AvCPR1zZq7R_xiXUG0hwhP5UxHA8Ix52JQp4",1787194072467]