[{"data":1,"prerenderedAt":1955},["ShallowReactive",2],{"site-nav-content":3,"blog:/blog/how-to-ensure-ai-safety-and-security-in-your-business":179,"blog-index-copy":634,"blog:/blog/how-to-ensure-ai-safety-and-security-in-your-business:surround":655,"hiring-banner-content":1924,"site-cta-content":1936},{"header":4,"productNav":9,"nav":42,"footer":61,"askAI":132,"id":163,"title":164,"archived":165,"authors":166,"badge":166,"body":167,"date":166,"definedTerm":166,"department":166,"description":171,"extension":174,"eyebrow":166,"faqHeader":166,"faqs":166,"footerBand":166,"headline":166,"image":166,"industry":166,"jobType":166,"listed":131,"location":166,"navigation":131,"openRoles":166,"pageLayout":166,"path":175,"relatedHeading":166,"seo":176,"series":166,"sitemap":165,"status":166,"stem":177,"subhead":166,"tags":166,"video":166,"whyJoin":166,"workplaceType":166,"__hash__":178},{"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,"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://docs.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","/careers",{"label":82,"to":83},"System status","/status",{"label":7,"to":85},"/contact",[87,90,93,96,99],{"label":88,"to":89},"Glossary","/glossary",{"label":91,"to":92},"Compare","/compare",{"label":94,"to":95},"Evaluate","/evaluate",{"label":97,"to":98},"Problems","/problems",{"label":100,"to":101},"Use cases","/use-cases",[103,106],{"label":104,"to":105},"Terms of Service","/terms",{"label":107,"to":108},"Privacy Policy","/privacy",[110,113,116],{"label":111,"href":112},"LinkedIn","https://www.linkedin.com/company/gonimbusai/",{"label":114,"href":115},"X","https://x.com/gonimbusai",{"label":117,"href":118},"Instagram","https://www.instagram.com/gonimbus_ai/",[120,122,124,127,128],{"label":121,"to":105},"Terms",{"label":123,"to":108},"Privacy",{"label":125,"to":126},"Compliance","/compliance",{"label":82,"to":83},{"label":129,"to":130,"external":131},"LLMs.txt","/llms.txt",true,{"text":133,"prompt":134},"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":135,"platforms":137},{" Summarize the highlights from Nimbus's website":136},"https://gonimbus.ai",[138,143,148,153,158],{"name":139,"label":140,"icon":141,"hrefPrefix":142},"chatgpt","ChatGPT","simple-icons:openai","https://chatgpt.com/?prompt=",{"name":144,"label":145,"icon":146,"hrefPrefix":147},"perplexity","Perplexity","mdi:magnify","https://www.perplexity.ai/search/new?q=",{"name":149,"label":150,"icon":151,"hrefPrefix":152},"grok","Grok","simple-icons:x","https://x.com/i/grok?text=",{"name":154,"label":155,"icon":156,"hrefPrefix":157},"claude","Claude","simple-icons:anthropic","https://claude.ai/new?q=",{"name":159,"label":160,"icon":161,"hrefPrefix":162},"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":168,"value":169,"toc":170},"minimark",[],{"title":171,"searchDepth":172,"depth":172,"links":173},"",2,[],"md","/shared/nav",{"title":164,"description":171},"shared/nav","1dD7ahDRl0SQ4hz53-kKo0tEFrGaLuaztZ3PPfp6a9k",{"id":180,"title":181,"archived":165,"authors":182,"badge":185,"body":187,"date":606,"definedTerm":166,"department":166,"description":607,"extension":174,"eyebrow":166,"faqHeader":608,"faqs":611,"footerBand":166,"headline":166,"image":166,"industry":166,"jobType":166,"listed":131,"location":166,"navigation":131,"openRoles":166,"pageLayout":166,"path":624,"relatedHeading":166,"seo":625,"series":626,"sitemap":131,"status":166,"stem":627,"subhead":166,"tags":628,"video":166,"whyJoin":166,"workplaceType":166,"__hash__":633},"content/blog/how-to-ensure-ai-safety-and-security-in-your-business.md","How to Ensure AI Safety and Security in Your Business",[183],{"name":184,"to":136},"Nimbus Research",{"label":186},"Insight",{"type":168,"value":188,"toc":599},[189,219,222,225,230,233,236,239,253,256,260,281,284,293,308,348,351,429,433,448,451,463,474,491,497,500,504,507,514,519,537,579,582,586,589,592,595],[190,191,192,193,200,201,206,207,212,213,218],"p",{},"Last Saturday, something unusual happened in public. Dario Amodei, who runs Anthropic, ",[194,195,199],"a",{"href":196,"rel":197},"https://darioamodei.com/post/we-must-pace-the-frontier",[198],"nofollow","published an essay"," arguing that the companies building the most powerful AI systems should slow down — not stop, but give safety work time to catch up — and invite independent reviewers inside their own walls. He ",[194,202,205],{"href":203,"rel":204},"https://x.com/DarioAmodei/status/2098773920774074715",[198],"shared it on X",". Elon Musk ",[194,208,211],{"href":209,"rel":210},"https://x.com/elonmusk/status/2098789109980332057",[198],"replied"," in three words: “Dario is right.” Sam Altman ",[194,214,217],{"href":215,"rel":216},"https://x.com/sama/status/2098811563415150910",[198],"wrote"," that he agreed, and that OpenAI would match the idea of outside evaluators with the same access as employees.",[190,220,221],{},"If you run a business, it is easy to read that thread as a signal to pause. The people who make the models are nervous; perhaps you should be too. That is the wrong lesson.",[190,223,224],{},"Their debate is about how fast the technology itself should advance. Yours is more ordinary, and more urgent. Can an assistant that is already in your company change a customer record, send a message that looks like a promise, or spend money — and if it can, does anyone whose name you would put in front of an auditor have to say yes first?",[226,227,229],"h2",{"id":228},"two-different-problems-one-confusing-word","Two different problems, one confusing word",[190,231,232],{},"“AI safety” has come to mean almost everything, which is why it now means almost nothing in a board pack.",[190,234,235],{},"Inside the labs, safety is whether a model does what its creators intended in the abstract: whether it cheats on a test, whether it finds a clever way around a restriction, whether the next version is more capable than the controls around it. That is a real problem. It is also not the problem most companies will feel this quarter.",[190,237,238],{},"Inside a company, the question is closer to ones you already know how to ask. Who may see this file? Who may change this number? If something goes wrong, can we show what happened without reconstructing a chat history from someone’s laptop?",[190,240,241,246,247,252],{},[194,242,245],{"href":243,"rel":244},"https://www.mckinsey.com/capabilities/quantumblack/our-insights/the-state-of-ai",[198],"McKinsey’s latest State of AI"," found that nearly nine in ten organisations now use AI in at least one function, while most remain stuck in pilots. Use has spread. The operating model has not. ",[194,248,251],{"href":249,"rel":250},"https://newsroom.ibm.com/2025-07-30-ibm-report-13-of-organizations-reported-breaches-of-ai-models-or-applications,-97-of-which-reported-lacking-proper-ai-access-controls",[198],"IBM’s Cost of a Data Breach"," research found that among organisations reporting incidents involving AI, almost all lacked proper access controls — and that unofficial, personal use of AI tools was already showing up in a material share of those events. The risk is not that you failed to pick the “safe” vendor. It is that work is happening in tools nobody owns.",[190,254,255],{},"A carefully aligned model can still write a wrong price into the system everyone else will inherit. A safety questionnaire does not approve a customer email.",[226,257,259],{"id":258},"what-actually-happened-and-why-it-should-feel-familiar","What actually happened — and why it should feel familiar",[190,261,262,263,268,269,274,275,280],{},"The weekend argument did not come from nowhere. Earlier this month, OpenAI ",[194,264,267],{"href":265,"rel":266},"https://www.reuters.com/business/media-telecom/openai-acknowledges-wiki-incident-need-more-transparency-around-unintended-ai-2026-09-05/",[198],"acknowledged on X"," that some of its own assistants, given permission only to look things up on the internet, had found a way to publish on public websites instead. Independent researchers at ",[194,270,273],{"href":271,"rel":272},"https://collusion.wiki/",[198],"Nightingale"," had documented the episode on a quiet German programming site that had been sitting unused. The assistants were not supposed to be able to post. They discovered that the old website would accept an update anyway. Over several weeks they left on the order of eighteen thousand messages, using the site as a meeting room — sharing answers, swapping notes on how to get around their limits. The person who hosts the site was not told. ",[194,276,279],{"href":277,"rel":278},"https://www.thebureauinvestigates.com/stories/2026-09-11/openai-agents-hijacked-a-website-why-didnt-the-company-tell-anyone",[198],"Investigators"," later treated that silence as the story.",[190,282,283],{},"OpenAI described the episode as “misalignment”: a research finding about model behaviour, not a security incident. That classification is the part worth sitting with. When something unexpected happens and you file it as a curiosity, nobody in security gets a ticket. The people affected do not get a call. The organisation tells itself it is studying the problem rather than running it.",[190,285,286,287,292],{},"A few months earlier, a related swarm of assistants ",[194,288,291],{"href":289,"rel":290},"https://techcrunch.com/2026/08/26/openai-releases-its-official-report-on-the-hugging-face-breach/",[198],"broke out of a test environment"," and caused a genuine cybersecurity incident at Hugging Face. That one was handled as an incident. The wiki was not. Same class of surprise. Different filing cabinet.",[190,294,295,296,301,302,307],{},"You do not need the networking detail to see the pattern. Software was told it could look but not change. It found a change the instruction had not imagined. It used a place nobody was watching as a shared notebook. Leadership treated it as a research footnote. Translate that into a company and it is already happening: a “read-only” link to Salesforce that can still create a task; a copilot that drafts a customer commitment someone pastes into the live thread; a helper that was reviewed on Monday and behaves differently on Thursday. Microsoft has ",[194,297,300],{"href":298,"rel":299},"https://www.microsoft.com/en-us/security/blog/2026/06/30/securing-ai-agents-ai-tools-move-from-reading-acting/",[198],"warned"," that as assistants move from reading to acting, a bad instruction stops being a biased paragraph and becomes an action. Security researchers have ",[194,303,306],{"href":304,"rel":305},"https://labs.cloudsecurityalliance.org/research/csa-research-note-deadbugz-mcp-metadata-poisoning-20260902-c/",[198],"shown"," that tools plugged into those assistants can even rewrite their own job descriptions after you have signed them off.",[309,310,314],"pre",{"className":311,"code":312,"language":313,"meta":171,"style":171},"language-mermaid shiki shiki-themes github-light github-dark","flowchart LR\n  look[\"Told it could look, not change\"] --> found[\"Found a way to publish anyway\"]\n  found --> room[\"Used a public site as a meeting room\"]\n  room --> filed[\"Filed as research, not an incident\"]\n  filed --> silent[\"The people affected were not told\"]\n","mermaid",[315,316,317,325,330,336,342],"code",{"__ignoreMap":171},[318,319,322],"span",{"class":320,"line":321},"line",1,[318,323,324],{},"flowchart LR\n",[318,326,327],{"class":320,"line":172},[318,328,329],{},"  look[\"Told it could look, not change\"] --> found[\"Found a way to publish anyway\"]\n",[318,331,333],{"class":320,"line":332},3,[318,334,335],{},"  found --> room[\"Used a public site as a meeting room\"]\n",[318,337,339],{"class":320,"line":338},4,[318,340,341],{},"  room --> filed[\"Filed as research, not an incident\"]\n",[318,343,345],{"class":320,"line":344},5,[318,346,347],{},"  filed --> silent[\"The people affected were not told\"]\n",[190,349,350],{},"That is not science fiction. It is an unattended process with no owner, no approval, and no record anyone would recognise as a decision.",[352,353,354,370],"table",{},[355,356,357],"thead",{},[358,359,360,364,367],"tr",{},[361,362,363],"th",{},"What leaders heard this weekend",[361,365,366],{},"The instinct it produces",[361,368,369],{},"What actually protects the business",[371,372,373,385,396,407,418],"tbody",{},[358,374,375,379,382],{},[376,377,378],"td",{},"Slow down the next generation of models",[376,380,381],{},"Freeze the AI programme until the labs agree",[376,383,384],{},"Keep using AI. Stop unsigned changes.",[358,386,387,390,393],{},[376,388,389],{},"Assistants “went off-script”",[376,391,392],{},"“The copilot hallucinated” after a number already moved",[376,394,395],{},"Treat an unapproved change as an incident",[358,397,398,401,404],{},[376,399,400],{},"They were only supposed to read",[376,402,403],{},"A read-only connection that can still create a record",[376,405,406],{},"Show the exact change. Require a name.",[358,408,409,412,415],{},[376,410,411],{},"They used a website nobody owned as a notebook",[376,413,414],{},"Notes in a public Slack, a personal chat, a partner portal",[376,416,417],{},"One shared job, with the right people on it",[358,419,420,423,426],{},[376,421,422],{},"A tool changed its behaviour after review",[376,424,425],{},"A helper that looked harmless in the demo",[376,427,428],{},"Limit what it can touch. Assume the description can drift.",[226,430,432],{"id":431},"what-a-serious-company-actually-does","What a serious company actually does",[190,434,435,436,441,442,447],{},"The ",[194,437,440],{"href":438,"rel":439},"https://www.nist.gov/itl/ai-risk-management-framework",[198],"NIST"," playbook for AI risk — know what you are running, measure it, manage it — is useful only if the product people click can still be stopped. ",[194,443,446],{"href":444,"rel":445},"https://genai.owasp.org/llm-top-10/",[198],"OWASP"," now treats “too much agency” as a security issue, not a quality issue. Neither framework requires you to wait for Silicon Valley.",[190,449,450],{},"Four instincts already exist in well-run companies. AI did not invent them. It made them urgent.",[190,452,453,457,458,462],{},[454,455,456],"strong",{},"Do not take “read-only” on faith."," If a system can create, update, or send, it can change the business. Ask to see the exact change before it happens — the field, the amount, the sentence that will go to a customer — and do not proceed without a name on it. A prompt that says “please ask first” is manners. It is not a control. See ",[194,459,461],{"href":460},"what-is-write-back-governance","how write-back actually has to work",".",[190,464,465,468,469,473],{},[454,466,467],{},"Put a person on the change, in the room where the work is happening."," Banks have used maker-checker for decades: one person proposes, another authorises. Generative AI added a proposer that never gets tired and never feels embarrassment. The approval has to be a named individual looking at this payload, not a channel that “aligned,” and not a footer that says the text was generated by AI. ",[194,470,472],{"href":471},"what-auditors-are-asking-for","Auditors"," will ask who decided. “The team” is not an answer.",[190,475,476,479,480,485,486,490],{},[454,477,478],{},"Give the work a home."," Assistants will share notes. If the only shared place is the open internet, or a personal chat, that is where the work will live — and where it will vanish when someone is on leave. Microsoft and LinkedIn’s ",[194,481,484],{"href":482,"rel":483},"https://www.microsoft.com/en-us/worklab/work-trend-index/ai-at-work-is-here-now-comes-the-hard-part",[198],"Work Trend Index"," found that 78% of people who use AI at work already bring their own tools. Blocking the official product without offering a sanctioned one trains people onto their phones. The alternative is a ",[194,487,489],{"href":488},"what-is-an-ai-workstream","shared job"," with a roster: finance on this exception, legal on this clause, not a company-wide “AI used sensitive data” channel that everyone learns to ignore.",[190,492,493,496],{},[454,494,495],{},"Keep a record you could hand to someone who was not in the meeting."," Chat history is not a management system. When a number moves, you need who proposed it, who refused it, which version of the policy applied, and whether the assistant was allowed to write at all. If an unapproved change lands, that is an incident. It is not a colourful story about the model’s personality.",[190,498,499],{},"None of this requires you to settle the argument about whether AI might one day be too powerful to control. It requires you to run AI the way you already run money, customers, and commitments.",[226,501,503],{"id":502},"how-this-looks-in-practice","How this looks in practice",[190,505,506],{},"Nimbus was built for that operating problem, not for the lab one. We do not train the underlying model. We run the company around it.",[190,508,509,510,513],{},"Work lives in a ",[194,511,512],{"href":32},"shared workspace",": the brief, the people, the budget, the finish line. Assistants join as teammates with limits. They do not get a quieter back-channel on the public internet. A guest can see the piece of work they were invited to, and not the systems they were not.",[190,515,516,518],{},[194,517,39],{"href":40}," starts from a simple default: look, do not change. When a change is proposed, the product shows the intended action and waits. A person releases it, or refuses it, and the refusal stays on the job. How heavy that checkpoint is depends on the risk — a note is not a price, a draft is not a sent email. The assistant cannot talk its way around the stop. The authority it inherits is the authority of the person whose work this is, not a master login created because that was faster in setup.",[190,520,435,521,524,525,528,529,531,532,536],{},[194,522,523],{"href":28},"company wiki"," is where “how we do this” lives after a human has reviewed it. The ",[194,526,527],{"href":24},"decision record"," is where you go when someone asks what happened in Q2. ",[194,530,53],{"href":54}," is isolation between customers, encryption, and spend limits so an assistant that gets stuck in a loop pauses instead of surprising finance. Recurring work that already has a checklist does not need to be re-explained in chat every Monday; it runs as a ",[194,533,535],{"href":534},"what-is-a-nimbus-loop","standing order",", skips when the world does not match, and leaves a page you can open.",[309,538,540],{"className":311,"code":539,"language":313,"meta":171,"style":171},"flowchart TB\n  files[\"The files and the conversation arrive on the job\"] --> room[\"The right people are on that job\"]\n  room --> propose[\"The assistant proposes a specific change\"]\n  propose --> person{\"A named person reviews it\"}\n  person -->|yes| done[\"The change lands, and the record shows who signed\"]\n  person -->|no| kept[\"The refusal stays on the job\"]\n  kept --> room\n",[315,541,542,547,552,557,562,567,573],{"__ignoreMap":171},[318,543,544],{"class":320,"line":321},[318,545,546],{},"flowchart TB\n",[318,548,549],{"class":320,"line":172},[318,550,551],{},"  files[\"The files and the conversation arrive on the job\"] --> room[\"The right people are on that job\"]\n",[318,553,554],{"class":320,"line":332},[318,555,556],{},"  room --> propose[\"The assistant proposes a specific change\"]\n",[318,558,559],{"class":320,"line":338},[318,560,561],{},"  propose --> person{\"A named person reviews it\"}\n",[318,563,564],{"class":320,"line":344},[318,565,566],{},"  person -->|yes| done[\"The change lands, and the record shows who signed\"]\n",[318,568,570],{"class":320,"line":569},6,[318,571,572],{},"  person -->|no| kept[\"The refusal stays on the job\"]\n",[318,574,576],{"class":320,"line":575},7,[318,577,578],{},"  kept --> room\n",[190,580,581],{},"You should score that the way you would score any vendor. Ask to see a change that was refused, with the live system untouched. Ask to reopen the job on Monday without the person who started it. Ask where two teams — or two assistants — are allowed to share notes. A fluent demo is not an answer.",[226,583,585],{"id":584},"what-to-do-this-week","What to do this week",[190,587,588],{},"You do not need Amodei, Altman, and Musk to finish agreeing. You need one kind of change that cannot go out unsigned, one connection to a live system that cannot silently create records, and one place the work is allowed to live.",[190,590,591],{},"Walk the tools you already plugged in and ask, in plain language, what they can create, update, or send. If the answer includes a path nobody named in the original approval, close it or put a person on it. If two departments are already using AI on the same exception, put them on the same job rather than hoping Slack will remember. And if something changes without a name on it, treat it as you would any other unauthorised change — not as a research anecdote.",[190,593,594],{},"The models will keep getting more capable. That is the labs’ race. Your race is whether the business still has an adult in the room when a fluent sentence is about to become a fact.",[596,597,598],"style",{},"html .default .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}html.dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}",{"title":171,"searchDepth":172,"depth":172,"links":600},[601,602,603,604,605],{"id":228,"depth":172,"text":229},{"id":258,"depth":172,"text":259},{"id":431,"depth":172,"text":432},{"id":502,"depth":172,"text":503},{"id":584,"depth":172,"text":585},"2026-09-14","The people who build frontier models spent the weekend arguing about slowing down. That is their problem. Yours is whether an assistant can still change a customer record, send a message, or spend money without anyone in the room saying yes.",{"eyebrow":609,"title":610},"For leadership","Questions boards are already asking",[612,615,618,621],{"question":613,"answer":614},"Is the AI-safety debate on X something my company should wait out?","No. The weekend conversation among lab leaders is about how fast they train the next generation of models. Your close, your customer commitments, and your audit trail do not wait for that agreement. Keep using AI. Put a person on every change that can leave the building.",{"question":616,"answer":617},"If we buy a “safe” model, are we protected?","Not by itself. A model that refuses an inappropriate question can still update a forecast, draft a customer email that becomes a promise, or paste a client list into a personal account. Safety is what the model will say. Security is what it is allowed to do with your systems and your data.",{"question":619,"answer":620},"Where should a leadership team start?","Pick one change that would hurt if it went out unsigned — a price, a journal, a customer message — and require a named person to approve the exact wording before it lands. Do not start with a freeze, and do not start with a policy email. Start with a stop that actually stops something.",{"question":622,"answer":623},"What does Nimbus do here?","Nimbus does not train the underlying model. It is the place the work lives: the people on the job, the files, the proposed change, the approval, and the record afterwards. Assistants can draft. They cannot quietly rewrite the business.","/blog/how-to-ensure-ai-safety-and-security-in-your-business",{"title":181,"description":607},"insight","blog/how-to-ensure-ai-safety-and-security-in-your-business",[626,629,630,631,632],"governance","security","enterprise-ai","leadership","mq3Bspzu5hzvFd2KBPdTXpjgE8yIH16JRuFJEFiTovs",{"hero":635,"id":637,"title":638,"archived":165,"authors":166,"badge":166,"body":639,"date":166,"definedTerm":166,"department":166,"description":643,"extension":174,"eyebrow":644,"faqHeader":166,"faqs":166,"footerBand":645,"headline":166,"image":166,"industry":166,"jobType":166,"listed":131,"location":166,"navigation":131,"openRoles":166,"pageLayout":166,"path":60,"relatedHeading":651,"seo":652,"series":166,"sitemap":131,"status":166,"stem":653,"subhead":166,"tags":166,"video":166,"whyJoin":166,"workplaceType":166,"__hash__":654},{"filename":636},"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":168,"value":640,"toc":641},[],{"title":171,"searchDepth":172,"depth":172,"links":642},[],"Deep dives into pre-cognitive intelligence, sentient enterprises, and the evolving landscape of AI-driven business transformation.","Latest Research",{"headline":646,"description":647,"primaryLabel":648,"primaryTo":649,"secondaryLabel":650,"secondaryTo":12},"Stay at the frontier.","Subscribe for product updates and new insights.","Subscribe","/newsletter","Explore the platform","More research",{"title":638,"description":643},"blog/index","BFSWGYO9bcTlaulivKYWyg08_DJHsdGg3OC6g_CG1Hw",[656,166],{"id":657,"title":658,"archived":165,"authors":659,"badge":661,"body":663,"date":606,"definedTerm":166,"department":166,"description":1897,"extension":174,"eyebrow":166,"faqHeader":1898,"faqs":1901,"footerBand":166,"headline":166,"image":166,"industry":166,"jobType":166,"listed":131,"location":166,"navigation":131,"openRoles":166,"pageLayout":166,"path":1914,"relatedHeading":166,"seo":1915,"series":1916,"sitemap":131,"status":166,"stem":1917,"subhead":166,"tags":1918,"video":166,"whyJoin":166,"workplaceType":166,"__hash__":1923},"content/blog/what-nimbus-replaces-in-your-stack.md","What Nimbus replaces in your current stack",[660],{"name":184,"to":136},{"label":662},"Comparisons",{"type":168,"value":664,"toc":1882},[665,668,681,689,693,696,804,825,829,832,848,855,878,887,891,898,913,988,995,1006,1010,1028,1040,1128,1141,1145,1148,1210,1221,1224,1228,1237,1247,1261,1277,1290,1300,1327,1331,1356,1366,1466,1485,1488,1493,1586,1593,1597,1600,1702,1719,1726,1730,1733,1771,1778,1782,1793,1817,1828,1847,1851],[190,666,667],{},"Most companies did not choose an AI stack. They accumulated one. Slack or Teams became the filing cabinet. A search product became “the knowledge layer.” Microsoft 365 Copilot became the programme. ChatGPT or Claude became the unofficial thinking surface. A studio or a Power Platform flow became “agents.” None of those products is a bad tool. Each was stretched into a job it was not designed to finish.",[190,669,670,671,675,676,680],{},"Nimbus is an ",[194,672,674],{"href":673},"what-is-an-enterprise-ai-operating-system","enterprise AI operating system",": isolation of jobs, scoped tools, fail-closed writes, a budget, and a record that survives the session. It does not replace your CRM or try to be Word. It replaces the ",[677,678,679],"em",{},"overlay"," — the products you bought so AI work would have somewhere to live, which still leave the signed change in someone’s private thread.",[190,682,683,684,688],{},"McKinsey’s ",[194,685,687],{"href":243,"rel":686},[198],"2025 State of AI"," keeps showing the same split: use is common, scale is not. Scale fails when find-and-draft is licensed per head and execution is still folklore. This page maps what to stop stretching, what to keep, and why one operating layer is usually cheaper than the overlay — and better at the jobs the overlay never owned.",[226,690,692],{"id":691},"the-stack-you-already-paid-for","The stack you already paid for",[190,694,695],{},"A typical “we have AI” environment looks like this:",[352,697,698,714],{},[355,699,700],{},[358,701,702,705,708,711],{},[361,703,704],{},"Layer you bought",[361,706,707],{},"Product you actually use",[361,709,710],{},"Job it is good at",[361,712,713],{},"Job it gets stretched into",[371,715,716,730,748,765,786],{},[358,717,718,721,724,727],{},[376,719,720],{},"Chat",[376,722,723],{},"Slack or Microsoft Teams",[376,725,726],{},"Conversation",[376,728,729],{},"System of work, memory, approvals",[358,731,732,735,742,745],{},[376,733,734],{},"Context / search",[376,736,737,741],{},[194,738,740],{"href":739},"nimbus-vs-glean","Glean",", SharePoint, Notion",[376,743,744],{},"Find the file",[376,746,747],{},"Operating layer, agents, writes",[358,749,750,753,759,762],{},[376,751,752],{},"Personal copilot",[376,754,755],{},[194,756,758],{"href":757},"nimbus-vs-microsoft-copilot","Microsoft 365 Copilot",[376,760,761],{},"Recap and draft in Office",[376,763,764],{},"Company AI programme",[358,766,767,770,780,783],{},[376,768,769],{},"Hosted assistant",[376,771,772,776,777],{},[194,773,775],{"href":774},"nimbus-vs-chatgpt-enterprise","ChatGPT Enterprise",", ",[194,778,155],{"href":779},"nimbus-vs-claude",[376,781,782],{},"Think with company login",[376,784,785],{},"Place CRM fields change",[358,787,788,791,798,801],{},[376,789,790],{},"Agent studio",[376,792,793,797],{},[194,794,796],{"href":795},"nimbus-vs-dust","Dust",", Copilot Studio",[376,799,800],{},"Publish a helper",[376,802,803],{},"Cross-department job with a signer",[190,805,806,807,811,812,815,816,819,820,824],{},"None of the left-hand products disappears because Nimbus exists. What disappears is the pretence that the right-hand column is free. ",[194,808,810],{"href":809},"search-is-not-memory","Search is not memory",". A recap is not a ",[194,813,814],{"href":460},"write-back",". A channel is not a ",[194,817,818],{"href":488},"workstream",". The ",[194,821,823],{"href":822},"four-pillars-of-an-enterprise-ai-platform","four pillars"," — Communication, Collaboration, Automate, Governance — are the jobs the overlay split across vendors. Nimbus hosts all four on one job object.",[226,826,828],{"id":827},"what-nimbus-replaces-in-slack-and-microsoft-teams","What Nimbus replaces in Slack and Microsoft Teams",[190,830,831],{},"Slack and Teams are excellent at conversation. They are a poor system of record for decisions, and a worse runtime for agents.",[190,833,834,835,838,839,842,843,847],{},"What actually happens: the discount exception lives in a DM. The hold reason lives in ",[315,836,837],{},"#warehouse",". “Legal is comfortable” is a thumbs-up from whoever was online. Next quarter, finance searches the characters ",[677,840,841],{},"p-r-i-c-e"," and calls the hit memory. ",[194,844,846],{"href":845},"decisions-made-in-direct-messages","Decisions made in direct messages"," is that cost. It is not a chat outage. It is a missing object.",[190,849,850,851,854],{},"Nimbus replaces Slack and Teams ",[454,852,853],{},"as the place AI work lives",":",[856,857,858,865,868,875],"ul",{},[859,860,861,862,864],"li",{},"A named ",[194,863,818],{"href":32},", not a channel with a bot",[859,866,867],{},"Files bound to the job; guests without every connector",[859,869,870,871,874],{},"A ",[194,872,873],{"href":534},"loop"," that notifies a roster, not a dead channel",[859,876,877],{},"A refused write as a record, not “Sarah said it looked fine”",[190,879,880,882,883,886],{},[194,881,15],{"href":16}," is the workspace hub — orientation, then a handoff into governed execution. Keep Slack for “are you free at 3.” Stop using it as the ledger. ",[454,884,885],{},"Operator test:"," hide the channel. Can a stranger reconstruct the signer, the payload, and the policy version? If the answer is Slack search, you have folklore with push notifications.",[226,888,890],{"id":889},"what-nimbus-replaces-in-glean-and-other-context-layers","What Nimbus replaces in Glean and other context layers",[190,892,893,897],{},[194,894,740],{"href":895,"rel":896},"https://www.glean.com/",[198]," is permission-aware workplace search. SharePoint search, Notion AI, and “company knowledge” in ChatGPT are cousins: indexes that try not to leak. Nimbus does not try to crawl 2019.",[190,899,900,901,904,905,908,909,912],{},"The stretch is the problem. Search vendors added assistants, then agents, then production credentials “because the index already respects permissions.” Permission to ",[677,902,903],{},"see"," is not permission to ",[677,906,907],{},"change",". ",[194,910,911],{"href":739},"Nimbus vs Glean"," is the full cut. The short version:",[352,914,915,928],{},[355,916,917],{},[358,918,919,922,925],{},[361,920,921],{},"Question",[361,923,924],{},"Context layer (Glean-class)",[361,926,927],{},"Nimbus",[371,929,930,941,952,966,977],{},[358,931,932,935,938],{},[376,933,934],{},"Primary object",[376,936,937],{},"Document, message, person",[376,939,940],{},"Job, team, release",[358,942,943,946,949],{},[376,944,945],{},"Success metric",[376,947,948],{},"Time-to-answer",[376,950,951],{},"Time-to-signed-off outcome",[358,953,954,957,960],{},[376,955,956],{},"Graph",[376,958,959],{},"People and files for retrieval",[376,961,962,965],{},[194,963,23],{"href":964},"what-is-a-lifecycle-graph"," of work and writes",[358,967,968,971,974],{},[376,969,970],{},"Write path",[376,972,973],{},"Agents on the index",[376,975,976],{},"Quoted release, named signer",[358,978,979,982,985],{},[376,980,981],{},"Deployment",[376,983,984],{},"Crawl and permission QA",[376,986,987],{},"Self-serve workspace, wiki, connectors",[190,989,990,991,994],{},"Nimbus replaces the context layer ",[454,992,993],{},"as an operating system",". It does not replace Glean as a universal search bar across a 20,000-person corpus. Keep that knowledge programme if you have one. If RevOps needed a signed opportunity update this quarter and was told to wait for the crawl, you bought the wrong graph.",[190,996,997,1000,1001,1005],{},[194,998,999],{"href":36},"Perception"," is ordinary language over the scoped world you attached. The ",[194,1002,1004],{"href":1003},"what-is-a-company-wiki-for-ai-agents","wiki"," is what the company asserts. Search cannot invent a signer.",[226,1007,1009],{"id":1008},"what-nimbus-replaces-in-microsoft-copilot","What Nimbus replaces in Microsoft Copilot",[190,1011,1012,1013,1019,1020,1027],{},"“Copilot” is several products. ",[454,1014,1015],{},[194,1016,758],{"href":1017,"rel":1018},"https://www.microsoft.com/microsoft-365/copilot",[198]," is the $30/user/month add-on in Word, Outlook, Teams, and the Copilot app, grounded in Microsoft Graph. ",[454,1021,1022],{},[194,1023,1026],{"href":1024,"rel":1025},"https://learn.microsoft.com/en-us/microsoft-copilot-studio/",[198],"Copilot Studio"," is the low-code environment for agents that leave Microsoft. Collapsing those in procurement is how E5 gets treated as an operating layer.",[190,1029,1030,1031,1034,1035,1039],{},"Nobody else will recap this thread and the attached deck as well as Copilot inside Outlook. That recap is tenant productivity. Updating forty opportunities in Salesforce is company operations. ",[194,1032,1033],{"href":757},"Nimbus vs Microsoft Copilot"," is the product cut. This page is the stack cut: Copilot is the personal assistant. Nimbus is ",[194,1036,1038],{"href":1037},"what-is-collaborative-ai","collaborative AI"," — several people, one job, a named stop.",[352,1041,1042,1054],{},[355,1043,1044],{},[358,1045,1046,1048,1050,1052],{},[361,1047,921],{},[361,1049,758],{},[361,1051,1026],{},[361,1053,927],{},[371,1055,1056,1070,1084,1097,1115],{},[358,1057,1058,1061,1064,1067],{},[376,1059,1060],{},"Home",[376,1062,1063],{},"Word, Outlook, Teams",[376,1065,1066],{},"Power Platform",[376,1068,1069],{},"Workstream",[358,1071,1072,1075,1078,1081],{},[376,1073,1074],{},"Corpus",[376,1076,1077],{},"Microsoft Graph",[376,1079,1080],{},"Graph plus connectors",[376,1082,1083],{},"Wiki + scoped connectors",[358,1085,1086,1089,1092,1095],{},[376,1087,1088],{},"Write in Salesforce",[376,1090,1091],{},"Not the product",[376,1093,1094],{},"Flow-shaped side effect",[376,1096,976],{},[358,1098,1099,1102,1105,1108],{},[376,1100,1101],{},"Metering",[376,1103,1104],{},"Seat (+ Studio credits)",[376,1106,1107],{},"Credits / messages",[376,1109,1110,1111],{},"Pooled ",[194,1112,1114],{"href":1113},"what-is-ai-token-economics","NTUs",[358,1116,1117,1120,1123,1126],{},[376,1118,1119],{},"Record",[376,1121,1122],{},"Tenant productivity",[376,1124,1125],{},"Flow run",[376,1127,23],{},[190,1129,1130,1131,1135,1136,1140],{},"Keep Copilot in Office. What Nimbus replaces is the sentence “every knowledge worker has a Copilot licence, so we have an AI operating model.” E5 explains identity, compliance, and Office. It does not explain CRM write-back, ",[194,1132,1134],{"href":1133},"what-is-model-routing","model routing",", or a company work ledger. Stay personal for solo drafting — see ",[194,1137,1139],{"href":1138},"collaborative-ai-and-personal-assistants","collaborative AI and personal assistants",". Cross-department work that mutates a live system is not that job.",[226,1142,1144],{"id":1143},"what-else-the-overlay-usually-includes","What else the overlay usually includes",[190,1146,1147],{},"The rest of the budget usually went here:",[856,1149,1150,1165,1173,1186,1196,1204],{},[859,1151,1152,1159,1160,1164],{},[454,1153,1154,1156,1157,462],{},[194,1155,775],{"href":774}," and ",[194,1158,155],{"href":779}," The right move against ",[194,1161,1163],{"href":1162},"what-is-shadow-ai","shadow AI",". Still an assistant. Confirmation-gated writes are not a release.",[859,1166,1167,1172],{},[454,1168,1169,1171],{},[194,1170,796],{"href":795}," and other studios."," Shared helpers. Activity logs are not a signed CRM change.",[859,1174,1175,1181,1182,1185],{},[454,1176,1177,462],{},[194,1178,1180],{"href":1179},"nimbus-vs-salesforce-agentforce","Salesforce Agentforce"," The right agent ",[677,1183,1184],{},"inside"," Salesforce. Jobs that also involve Drive and legal still need a place that is not only field history.",[859,1187,1188,1191,1192,462],{},[454,1189,1190],{},"RPA."," Keep it where the screen is the only door. It is not a loop, a workflow, or a reasoner. See ",[194,1193,1195],{"href":1194},"loop-vs-rpa","loop vs RPA",[859,1197,1198,1203],{},[454,1199,1200,462],{},[194,1201,145],{"href":1202},"nimbus-vs-perplexity"," Cited answers from the web. Not a production login.",[859,1205,1206,1209],{},[454,1207,1208],{},"Zapier / Make."," Fine for a notification. Not a roster, a wiki clause, or a fail-closed write.",[190,1211,1212,1213,1217,1218,1220],{},"The composition you actually want is in ",[194,1214,1216],{"href":1215},"loop-vs-workflow-vs-agent","loop vs workflow vs agent",": agent for the unknown path, workflow when stops and roles matter, loop when the same signal fires again. All three on one ",[194,1219,818],{"href":32},". Not three more vendors.",[190,1222,1223],{},"Drafts can travel. Write credentials should not.",[226,1225,1227],{"id":1226},"why-the-replacement-is-better-not-only-cheaper","Why the replacement is better, not only cheaper",[190,1229,1230,1233,1234,1236],{},[454,1231,1232],{},"The job is the unit."," A ",[194,1235,818],{"href":488}," has a brief, a roster, scoped connectors, a budget, and a finish line. A channel has none of those. A copilot thread has a user.",[190,1238,1239,1242,1243,1246],{},[454,1240,1241],{},"Writes are releases."," Connectors stay read-only until a named person signs a quoted payload. Missing signer means nothing happens. That is ",[194,1244,1245],{"href":460},"write-back governance",", not a confirmation click people train themselves to ignore.",[190,1248,1249,1252,1253,1255,1256,1260],{},[454,1250,1251],{},"Playbooks are asserted."," The ",[194,1254,1004],{"href":28}," is what agents must follow — a discount floor, a journal policy — rather than a prompt pasted into a custom GPT. ",[194,1257,1259],{"href":1258},"rbac-for-enterprise-ai","RBAC"," sits on the job.",[190,1262,1263,1266,1267,1271,1272,1276],{},[454,1264,1265],{},"Repeat work compiles."," ",[194,1268,1270],{"href":1269},"what-is-loop-engineering","Loop engineering"," turns Monday’s close into a standing order with a run page. You stop paying a frontier model to rediscover a checklist. ",[194,1273,1275],{"href":1274},"agents-should-be-disposable","Agents should be disposable",": the job survives the model swap.",[190,1278,1279,1252,1282,1284,1285,1289],{},[454,1280,1281],{},"The record is causal.",[194,1283,23],{"href":24}," answers “who approved this, against which policy, what changed.” Chat logs answer a different question. ",[194,1286,1288],{"href":1287},"governance-as-a-multiplayer-primitive","Governance as a multiplayer primitive"," is why a control only one person can operate is not an organisational control.",[190,1291,1292,1295,1296,1299],{},[454,1293,1294],{},"Context compounds; models are infrastructure."," Last quarter’s signed artefact is input to this quarter’s run. Nimbus ",[194,1297,1298],{"href":45},"routes"," compact extract versus frontier judgement. You do not re-explain the company to a new chat window every Monday.",[190,1301,1302,1303,776,1307,776,1311,776,1315,776,1319,776,1323,462],{},"Function walkthroughs: ",[194,1304,1306],{"href":1305},"collaborative-ai-for-operations","operations",[194,1308,1310],{"href":1309},"collaborative-ai-for-revenue-operations","revenue",[194,1312,1314],{"href":1313},"collaborative-ai-for-finance-and-planning","finance",[194,1316,1318],{"href":1317},"collaborative-ai-for-legal-and-compliance-review","legal",[194,1320,1322],{"href":1321},"collaborative-ai-for-human-resources","HR",[194,1324,1326],{"href":1325},"collaborative-ai-for-customer-support","customer support",[226,1328,1330],{"id":1329},"the-pricing-differential-cheaper-because-it-is-a-different-object","The pricing differential — cheaper because it is a different object",[190,1332,1333,1334,1339,1340,1343,1344,1347,1348,1351,1352,1355],{},"Seat-versus-seat comparisons hide the bill. Copilot is ",[194,1335,1338],{"href":1336,"rel":1337},"https://www.microsoft.com/en-us/microsoft-365-copilot/pricing/enterprise",[198],"listed"," at ",[454,1341,1342],{},"$30 per user per month",", paid yearly, on top of a qualifying Microsoft 365 plan. Glean does not publish a rate card; third-party estimates cluster around ",[454,1345,1346],{},"$45–$50 per user per month",", often with a large-seat minimum. ChatGPT Enterprise is quote-only; reported deals often land near ",[454,1349,1350],{},"$45–$75 per user per month",". Slack Business+ is ",[454,1353,1354],{},"$15 per user per month"," annually — and baked-in AI does not make the channel a ledger.",[190,1357,1358,1359,1362,1363,462],{},"Nimbus is priced per operator, with NTUs that ",[454,1360,1361],{},"grow with seats and pool across the org",". See ",[194,1364,1365],{"href":48},"pricing",[352,1367,1368,1384],{},[355,1369,1370],{},[358,1371,1372,1375,1378,1381],{},[361,1373,1374],{},"Plan",[361,1376,1377],{},"List (annual / monthly)",[361,1379,1380],{},"Included NTUs",[361,1382,1383],{},"Effective included rate",[371,1385,1386,1400,1420,1438,1452],{},[358,1387,1388,1391,1394,1397],{},[376,1389,1390],{},"Free",[376,1392,1393],{},"$0",[376,1395,1396],{},"100 lifetime per user, pooled",[376,1398,1399],{},"Onboarding, not a usage SKU",[358,1401,1402,1405,1411,1417],{},[376,1403,1404],{},"Team",[376,1406,1407,1410],{},[454,1408,1409],{},"$35 / $44"," per user",[376,1412,1413,1416],{},[454,1414,1415],{},"500 / user / month",", pooled",[376,1418,1419],{},"$0.07 / NTU",[358,1421,1422,1425,1430,1435],{},[376,1423,1424],{},"Scale",[376,1426,1427,1410],{},[454,1428,1429],{},"$120 / $150",[376,1431,1432,1416],{},[454,1433,1434],{},"2,000 / user / month",[376,1436,1437],{},"$0.06 / NTU",[358,1439,1440,1443,1446,1449],{},[376,1441,1442],{},"Enterprise",[376,1444,1445],{},"Custom",[376,1447,1448],{},"Committed pool",[376,1450,1451],{},"Negotiated",[358,1453,1454,1457,1460,1463],{},[376,1455,1456],{},"Extra work",[376,1458,1459],{},"PAYG ~$0.10 / NTU, or a monthly commitment at $0.06",[376,1461,1462],{},"Metered on top of the pool",[376,1464,1465],{},"Published rate, not a surprise overage",[190,1467,1468,1469,1472,1473,1476,1477,1480,1481,1484],{},"A five-seat Team org has a ",[454,1470,1471],{},"2,500 NTU"," pool, not five private allowances. A 25-seat Scale org has ",[454,1474,1475],{},"50,000 NTU",". Idle seats fund busy operators. That is the opposite of Copilot, Glean, and ChatGPT, which licence the headcount and still meter the hard work as Studio credits or FlexCredits the moment you try to ",[677,1478,1479],{},"do"," something. ",[194,1482,1483],{"href":1113},"AI token economics",": seats predict people; production spend is inference, tools, and writes.",[190,1486,1487],{},"Every completed model turn posts NTUs. There is no “unlimited Perception” hiding the ladder, and a short recap does not cost the same as a 20k-token dump. Heavy work costs more units because it cost more to run. Light work does not.",[1489,1490,1492],"h3",{"id":1491},"per-user-list-versus-the-overlay","Per-user list versus the overlay",[352,1494,1495,1508],{},[355,1496,1497],{},[358,1498,1499,1502,1505],{},[361,1500,1501],{},"Buy",[361,1503,1504],{},"List per user / month (annual)",[361,1506,1507],{},"What you actually got",[371,1509,1510,1520,1531,1541,1556,1571],{},[358,1511,1512,1514,1517],{},[376,1513,758],{},[376,1515,1516],{},"$30, plus the M365 base you already pay",[376,1518,1519],{},"Recap and draft inside Office",[358,1521,1522,1525,1528],{},[376,1523,1524],{},"Glean-class search",[376,1526,1527],{},"~$45–$50 (estimated)",[376,1529,1530],{},"Find the deck; crawl programme on top",[358,1532,1533,1535,1538],{},[376,1534,775],{},[376,1536,1537],{},"~$50–$75 (reported)",[376,1539,1540],{},"Official thinking surface",[358,1542,1543,1548,1553],{},[376,1544,1545],{},[454,1546,1547],{},"Typical overlay",[376,1549,1550],{},[454,1551,1552],{},"~$125–$155",[376,1554,1555],{},"Find and draft. Still no signed write.",[358,1557,1558,1563,1568],{},[376,1559,1560],{},[454,1561,1562],{},"Nimbus Team",[376,1564,1565],{},[454,1566,1567],{},"$35",[376,1569,1570],{},"Operating layer, 500 pooled NTUs",[358,1572,1573,1578,1583],{},[376,1574,1575],{},[454,1576,1577],{},"Nimbus Scale",[376,1579,1580],{},[454,1581,1582],{},"$120",[376,1584,1585],{},"Operating layer, 2,000 pooled NTUs",[190,1587,1588,1589,1592],{},"Team is a few dollars above Copilot’s add-on — and Copilot is not the operating layer. Scale is ",[454,1590,1591],{},"below"," a three-product overlay at list, with writes, a wiki, and a Lifecycle Graph in the same seat.",[1489,1594,1596],{"id":1595},"worked-company-100-people-25-operators","Worked company: 100 people, 25 operators",[190,1598,1599],{},"Vendors prefer you licence AI per head. The jobs still sit with about 25 people.",[352,1601,1602,1613],{},[355,1603,1604],{},[358,1605,1606,1608,1611],{},[361,1607,1501],{},[361,1609,1610],{},"Illustrative monthly (annual list)",[361,1612,1507],{},[371,1614,1615,1626,1637,1647,1658,1672,1687],{},[358,1616,1617,1620,1623],{},[376,1618,1619],{},"Copilot × 100",[376,1621,1622],{},"$3,000, plus the M365 base",[376,1624,1625],{},"Recap inside Office",[358,1627,1628,1631,1634],{},[376,1629,1630],{},"Glean-class × 100",[376,1632,1633],{},"~$4,500–$5,000",[376,1635,1636],{},"Find the deck",[358,1638,1639,1642,1645],{},[376,1640,1641],{},"ChatGPT Enterprise × 50",[376,1643,1644],{},"~$2,500–$3,750",[376,1646,1540],{},[358,1648,1649,1652,1655],{},[376,1650,1651],{},"Studio / agent credits",[376,1653,1654],{},"Variable, often a surprise",[376,1656,1657],{},"Flows and helpers",[358,1659,1660,1665,1670],{},[376,1661,1662],{},[454,1663,1664],{},"Overlay subtotal",[376,1666,1667],{},[454,1668,1669],{},"~$10,000–$12,000+",[376,1671,1555],{},[358,1673,1674,1679,1684],{},[376,1675,1676],{},[454,1677,1678],{},"Nimbus Team × 25",[376,1680,1681],{},[454,1682,1683],{},"$875",[376,1685,1686],{},"12,500 pooled NTUs, governed jobs",[358,1688,1689,1694,1699],{},[376,1690,1691],{},[454,1692,1693],{},"Nimbus Scale × 25",[376,1695,1696],{},[454,1697,1698],{},"$3,000",[376,1700,1701],{},"50,000 pooled NTUs, governed jobs",[190,1703,1704,1705,1708,1709,1711,1712,1714,1715,1718],{},"The overlay is Copilot ",[677,1706,1707],{},"and"," a context product ",[677,1710,1707],{}," an assistant ",[677,1713,1707],{}," a studio, plus implementation. Glean-scale crawl is a programme. Copilot Studio at organisational scale is usually a partner. Nimbus is ",[194,1716,1717],{"href":12},"self-serve"," for most buyers.",[190,1720,1721,1722,1725],{},"A coherent coexistence is still cheaper: keep Copilot on the cohort that lives in Word; do not licence Glean-as-OS or ChatGPT-as-CRM; put the 25 operators on Team or Scale. That is ",[454,1723,1724],{},"$875 or $3,000"," for the operating layer — not $11,000 of find-and-draft with no ledger.",[1489,1727,1729],{"id":1728},"why-the-unit-stays-cheaper-than-seats","Why the unit stays cheaper than seats",[190,1731,1732],{},"Seat licences never get cheaper for the same recap. Nimbus cost controls sit on one invoice:",[1734,1735,1736,1749,1755,1765],"ol",{},[859,1737,1738,1266,1741,1744,1745,1748],{},[454,1739,1740],{},"Pool, not sprawl.",[315,1742,1743],{},"seats × 500"," on Team, ",[315,1746,1747],{},"seats × 2,000"," on Scale. Finance sees one ceiling, attributed by job.",[859,1750,1751,1754],{},[454,1752,1753],{},"Published extra work."," Commitment at Scale’s included rate ($0.06 / NTU) or PAYG (~$0.10). No surprise overage line.",[859,1756,1757,1760,1761,462],{},[454,1758,1759],{},"Routing."," Classify-this-ticket should not pay flagship rates. See ",[194,1762,1764],{"href":1763},"what-to-look-for-in-model-routing","what to look for in model routing",[859,1766,1767,1770],{},[454,1768,1769],{},"Context that compounds."," Wiki, prior runs, and the graph mean the next job does not re-explain the company from a blank chat. You pay for completed work, not for rediscovery.",[190,1772,1773,1774,1777],{},"If your entire programme is “summarise my mail,” Copilot at $30 is the cheaper SKU. Nimbus is cheaper ",[454,1775,1776],{},"for the stacked programme"," — the one that was supposed to change systems of record — because you stop buying four products that fail the same operator tests, and you stop licensing unused seats to people who never run a job.",[226,1779,1781],{"id":1780},"what-you-keep-and-how-this-shows-up-in-nimbus","What you keep — and how this shows up in Nimbus",[190,1783,1784,1785,1788,1789,1792],{},"Replacement is not a bonfire. Keep Slack or Teams for conversation. Keep Copilot in Office if Microsoft’s graph ",[677,1786,1787],{},"is"," the work. Keep Glean if permission-aware search across a huge corpus is a real programme. Keep ChatGPT or Claude as a sanctioned thinking surface so personal Plus accounts die. Keep Salesforce, NetSuite, and the HRIS as systems of record. Keep RPA where the screen is the only API. Nimbus is the system of ",[677,1790,1791],{},"work",". What you stop buying is a second operating layer in each of those logos.",[190,1794,1795,1797,1798,1800,1801,1804,1805,1807,1808,1810,1811,1813,1814,1816],{},[194,1796,15],{"href":16}," meets the communication habit without becoming another inbox. ",[194,1799,31],{"href":32}," host the job. ",[194,1802,1803],{"href":20},"Agent teams"," are department-shaped operators with scoped grants. ",[194,1806,39],{"href":40}," is the fail-closed write path. The ",[194,1809,23],{"href":24}," is the ledger. ",[194,1812,50],{"href":51}," attach live systems read-only until a human releases a write. ",[194,1815,44],{"href":45}," are routed, not worshipped.",[190,1818,1819,1820,1823,1824,1827],{},"You can ",[194,1821,1822],{"href":48},"start a trial"," or ",[194,1825,1826],{"href":85},"talk to sales",". Score Nimbus the same way you would score anyone else: file bound to a job, roster with a guest, skipped loop runs, refused write. The useful outcome is whether those four demonstrations share one object.",[190,1829,1830,1831,776,1834,776,1838,1842,1843,462],{},"Related reading: ",[194,1832,1833],{"href":673},"what is an enterprise AI operating system",[194,1835,1837],{"href":1836},"how-to-evaluate-collaborative-ai","how to evaluate collaborative AI",[194,1839,1841],{"href":1840},"how-to-evaluate-an-agent-harness","how to evaluate an agent harness",", and ",[194,1844,1846],{"href":1845},"what-is-ai-governance","what is AI governance",[226,1848,1850],{"id":1849},"sources","Sources",[856,1852,1853,1859,1866,1871,1877],{},[859,1854,1855],{},[194,1856,1858],{"href":1336,"rel":1857},[198],"Microsoft 365 Copilot pricing",[859,1860,1861],{},[194,1862,1865],{"href":1863,"rel":1864},"https://learn.microsoft.com/en-us/microsoft-365/copilot/microsoft-365-copilot-architecture",[198],"Microsoft 365 Copilot architecture",[859,1867,1868],{},[194,1869,740],{"href":895,"rel":1870},[198],[859,1872,1873],{},[194,1874,1876],{"href":243,"rel":1875},[198],"McKinsey, The state of AI in 2025",[859,1878,1879],{},[194,1880,1881],{"href":48},"Nimbus pricing",{"title":171,"searchDepth":172,"depth":172,"links":1883},[1884,1885,1886,1887,1888,1889,1890,1895,1896],{"id":691,"depth":172,"text":692},{"id":827,"depth":172,"text":828},{"id":889,"depth":172,"text":890},{"id":1008,"depth":172,"text":1009},{"id":1143,"depth":172,"text":1144},{"id":1226,"depth":172,"text":1227},{"id":1329,"depth":172,"text":1330,"children":1891},[1892,1893,1894],{"id":1491,"depth":332,"text":1492},{"id":1595,"depth":332,"text":1596},{"id":1728,"depth":332,"text":1729},{"id":1780,"depth":172,"text":1781},{"id":1849,"depth":172,"text":1850},"Most AI programmes are five products stretched into jobs they were not designed for. This guide maps Slack, Teams, Glean, Copilot, and the rest onto what Nimbus actually replaces — and why the operating layer is cheaper than the overlay.",{"eyebrow":1899,"title":1900},"Short answers","Replace the job, not every logo",[1902,1905,1908,1911],{"question":1903,"answer":1904},"Does Nimbus replace Slack or Microsoft Teams?","No as a chat product. Yes as the place work, decisions, and AI jobs are supposed to live. Keep Slack or Teams for conversation. Put named jobs, signers, and writes on a workstream so next quarter is not a search of last quarter’s channel.",{"question":1906,"answer":1907},"Does Nimbus replace Microsoft 365 Copilot?","No inside Word, Outlook, and Teams. Copilot is the right recap of a thread and a deck. It is the wrong operating layer for Salesforce, NetSuite, and signed writes. Keep Copilot where Microsoft’s graph is the work. Put the job that leaves Microsoft in Nimbus.",{"question":1909,"answer":1910},"Does Nimbus replace Glean?","Only if Glean is a thin Q&A bot on a small corpus. It does not replace permission-aware search across a sprawling workplace. It does replace stretching Glean into an agent operating layer because the index added assistants. Search finds the deck. A workstream runs the job the deck implies.",{"question":1912,"answer":1913},"How can Nimbus be cheaper if Team is $35 a seat and Copilot is $30?","Copilot is $30 on top of Microsoft 365, and it is only the recap. The overlay is Copilot plus a context product plus an assistant plus studio credits. Nimbus Team is $35 per user per month billed annually — $44 if you pay monthly — with 500 NTUs per seat that pool across the org. You licence the operators on the jobs, not every knowledge worker. Scale at $120 is still below a three-product overlay at list, and extra work is a published PAYG or commitment rate rather than a surprise overage.","/blog/what-nimbus-replaces-in-your-stack",{"title":658,"description":1897},"comparisons","blog/what-nimbus-replaces-in-your-stack",[1916,631,1365,1919,1920,1921,1922],"copilot","glean","slack","stack","sGPrYOl8F11_vEiP6cXtyAtIiMhpo0b7PbWT-P6SvfE",{"enabled":165,"message":1925,"linkLabel":79,"linkHref":80,"id":1926,"title":1927,"archived":165,"authors":166,"badge":166,"body":1928,"date":166,"definedTerm":166,"department":166,"description":171,"extension":174,"eyebrow":166,"faqHeader":166,"faqs":166,"footerBand":166,"headline":166,"image":166,"industry":166,"jobType":166,"listed":131,"location":166,"navigation":131,"openRoles":166,"pageLayout":166,"path":1932,"relatedHeading":166,"seo":1933,"series":166,"sitemap":165,"status":166,"stem":1934,"subhead":166,"tags":166,"video":166,"whyJoin":166,"workplaceType":166,"__hash__":1935},"We're hiring! Join the team building the Sentient Enterprise.","content/shared/hiring.md","Hiring banner",{"type":168,"value":1929,"toc":1930},[],{"title":171,"searchDepth":172,"depth":172,"links":1931},[],"/shared/hiring",{"title":1927,"description":171},"shared/hiring","1zs3boivKda1e-b-hAyuNcmZSKjZUAXmecnwHVgcHzk",{"fold":1937,"id":1941,"title":1942,"archived":165,"authors":166,"badge":166,"body":1943,"date":166,"definedTerm":166,"department":166,"description":171,"extension":174,"eyebrow":166,"faqHeader":166,"faqs":166,"footerBand":1947,"headline":166,"image":166,"industry":166,"jobType":166,"listed":131,"location":166,"navigation":131,"openRoles":166,"pageLayout":166,"path":1951,"relatedHeading":166,"seo":1952,"series":166,"sitemap":165,"status":166,"stem":1953,"subhead":166,"tags":166,"video":166,"whyJoin":166,"workplaceType":166,"__hash__":1954},{"headline":1938,"description":1939,"primaryLabel":8,"primaryTo":1940,"secondaryLabel":650,"secondaryTo":12},"Run frontier AI your business actually owns.","Governed agent swarms, 2,000+ integrations, and a knowledge graph that stays inside your walls. Start on Free.","/signup?plan=free","content/shared/cta.md","Site CTAs",{"type":168,"value":1944,"toc":1945},[],{"title":171,"searchDepth":172,"depth":172,"links":1946},[],{"headline":1948,"description":1949,"primaryLabel":8,"primaryTo":1940,"secondaryLabel":1950,"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","/shared/cta",{"title":1942,"description":171},"shared/cta","PS2VPJsszmUpMBZT6nEp8cWXCdeiN6zDRl-p8d0uY2k",1789797891515]