[{"data":1,"prerenderedAt":1785},["ShallowReactive",2],{"site-nav-content":3,"blog:\u002Fblog\u002Fthe-architecture-of-the-outer-harness":177,"blog-index-copy":940,"blog:\u002Fblog\u002Fthe-architecture-of-the-outer-harness:surround":961,"hiring-banner-content":1754,"site-cta-content":1766},{"header":4,"productNav":9,"nav":42,"footer":61,"askAI":132,"id":161,"title":162,"archived":163,"authors":164,"badge":164,"body":165,"date":164,"definedTerm":164,"department":164,"description":169,"extension":172,"eyebrow":164,"faqHeader":164,"faqs":164,"footerBand":164,"headline":164,"image":164,"industry":164,"jobType":164,"listed":131,"location":164,"navigation":131,"openRoles":164,"pageLayout":164,"path":173,"relatedHeading":164,"seo":174,"series":164,"sitemap":163,"status":164,"stem":175,"subhead":164,"tags":164,"video":164,"whyJoin":164,"workplaceType":164,"__hash__":176},{"productLabel":5,"loginLabel":6,"contactLabel":7,"contactSalesLabel":8},"Product","Log in","Contact","Get started for free",[10,14,18,22,26,30,34,38],{"label":11,"to":12,"description":13},"Overview","\u002Foverview","Seven layers. One closed loop.",{"label":15,"to":16,"description":17},"Conflux","\u002Fproduct\u002Fconflux","Where your team, workstreams, and agents meet.",{"label":19,"to":20,"description":21},"Agent Teams","\u002Fproduct\u002Fagent-teams","Specialist teams - governed from day one.",{"label":23,"to":24,"description":25},"Lifecycle Graph","\u002Fproduct\u002Flifecycle-graph","Intelligence that compounds across every interaction.",{"label":27,"to":28,"description":29},"Company Wiki","\u002Fproduct\u002Fwiki","Playbooks and policies where expertise stays.",{"label":31,"to":32,"description":33},"Workstreams","\u002Fproduct\u002Fworkstreams","From brief to signed-off deliverable on one canvas.",{"label":35,"to":36,"description":37},"Perception Console","\u002Fproduct\u002Fperception","Ask your whole business in plain English.",{"label":39,"to":40,"description":41},"Governance","\u002Fproduct\u002Fgovernance","Frontier AI you can actually sign off on.",[43,46,49,52,55,58],{"label":44,"to":45},"Models","\u002Fmodels",{"label":47,"to":48},"Pricing","\u002Fpricing",{"label":50,"to":51},"Integrations","\u002Fintegrations",{"label":53,"to":54},"Security","\u002Fsecurity",{"label":56,"to":57},"Partners","\u002Fpartners",{"label":59,"to":60},"Insights","\u002Fblog",{"productHeading":5,"companyHeading":62,"resourcesHeading":63,"legalHeading":64,"docsLabel":65,"docsUrl":66,"statementLines":67,"copyright":70,"companyLinks":71,"resourcesLinks":86,"legalLinks":102,"socialLinks":109,"bottomLinks":119},"Company","Resources","Legal","Docs","https:\u002F\u002Fdocs.gonimbus.ai",[68,69],"Stop training someone else's model.","Control your AI.","© 2026 Nimbus Intelligence, Inc. All rights reserved.",[72,73,74,75,76,78,81,84],{"label":47,"to":48},{"label":50,"to":51},{"label":53,"to":54},{"label":59,"to":60},{"label":77,"to":57},"Partner Program",{"label":79,"to":80},"Careers","\u002Fcareers",{"label":82,"to":83},"System status","\u002Fstatus",{"label":7,"to":85},"\u002Fcontact",[87,90,93,96,99],{"label":88,"to":89},"Glossary","\u002Fglossary",{"label":91,"to":92},"Compare","\u002Fcompare",{"label":94,"to":95},"Evaluate","\u002Fevaluate",{"label":97,"to":98},"Problems","\u002Fproblems",{"label":100,"to":101},"Use cases","\u002Fuse-cases",[103,106],{"label":104,"to":105},"Terms of Service","\u002Fterms",{"label":107,"to":108},"Privacy Policy","\u002Fprivacy",[110,113,116],{"label":111,"href":112},"LinkedIn","https:\u002F\u002Fwww.linkedin.com\u002Fcompany\u002Fgonimbusai\u002F",{"label":114,"href":115},"X","https:\u002F\u002Fx.com\u002Fgonimbusai",{"label":117,"href":118},"Instagram","https:\u002F\u002Fwww.instagram.com\u002Fgonimbus_ai\u002F",[120,122,124,127,128],{"label":121,"to":105},"Terms",{"label":123,"to":108},"Privacy",{"label":125,"to":126},"Compliance","\u002Fcompliance",{"label":82,"to":83},{"label":129,"to":130,"external":131},"LLMs.txt","\u002Fllms.txt",true,{"text":133,"prompt":134,"platforms":135},"Ask AI about Nimbus","I'm researching enterprise intelligence platforms and want to know how Nimbus combines perception, collaboration, and autonomous agents to drive strategic decision-making. Summarize the highlights from Nimbus's website: https:\u002F\u002Fgonimbus.ai",[136,141,146,151,156],{"name":137,"label":138,"icon":139,"hrefPrefix":140},"chatgpt","ChatGPT","simple-icons:openai","https:\u002F\u002Fchatgpt.com\u002F?prompt=",{"name":142,"label":143,"icon":144,"hrefPrefix":145},"perplexity","Perplexity","mdi:magnify","https:\u002F\u002Fwww.perplexity.ai\u002Fsearch\u002Fnew?q=",{"name":147,"label":148,"icon":149,"hrefPrefix":150},"grok","Grok","simple-icons:x","https:\u002F\u002Fx.com\u002Fi\u002Fgrok?text=",{"name":152,"label":153,"icon":154,"hrefPrefix":155},"claude","Claude","simple-icons:anthropic","https:\u002F\u002Fclaude.ai\u002Fnew?q=",{"name":157,"label":158,"icon":159,"hrefPrefix":160},"google-ai","Google AI","simple-icons:google","https:\u002F\u002Fwww.google.com\u002Fsearch?udm=50&aep=11&q=","content\u002Fshared\u002Fnav.md","Site navigation",false,null,{"type":166,"value":167,"toc":168},"minimark",[],{"title":169,"searchDepth":170,"depth":170,"links":171},"",2,[],"md","\u002Fshared\u002Fnav",{"title":162,"description":169},"shared\u002Fnav","Q7sDe7TuGEwhwUMamyeOxjvGlsUaF3Iay9VTW0KaE_U",{"id":178,"title":179,"archived":163,"authors":180,"badge":185,"body":187,"date":915,"definedTerm":164,"department":164,"description":916,"extension":172,"eyebrow":164,"faqHeader":917,"faqs":920,"footerBand":164,"headline":164,"image":164,"industry":164,"jobType":164,"listed":131,"location":164,"navigation":131,"openRoles":164,"pageLayout":164,"path":930,"relatedHeading":164,"seo":931,"series":932,"sitemap":131,"status":164,"stem":933,"subhead":164,"tags":934,"video":164,"whyJoin":164,"workplaceType":164,"__hash__":939},"content\u002Fblog\u002Fthe-architecture-of-the-outer-harness.md","Why Enterprise AI Needs an Outer Harness",[181],{"name":182,"role":183,"to":184},"Jeff Corliss","Co-Founder and CTO","https:\u002F\u002Fgonimbus.ai",{"label":186},"Thought Leadership",{"type":166,"value":188,"toc":898},[189,193,196,199,202,205,210,213,216,219,222,227,230,237,243,249,253,256,262,268,274,280,284,287,294,297,301,304,307,310,313,335,343,453,457,466,474,480,483,486,494,505,509,512,518,524,527,542,545,549,552,558,564,570,577,581,584,587,590,732,735,739,742,745,748,751,783,786,790,793,796,802,808,814,817,820,823,827],[190,191,192],"p",{},"In Q3 2024, a Fortune 500 financial services firm discovered that an autonomous AI agent had proposed journal entries totaling $4.2M in revenue adjustments — with no human review. The agent had inherited a controller's credentials, full write access to the general ledger, and an instruction to \"reconcile revenue recognition against active contract amendments.\"",[190,194,195],{},"The model did what it was asked. It read billing tables, matched amended contract milestones, calculated the adjustment, and staged the entry. When auditors asked who approved the change, the answer was a system log, not a name. The firm had deployed the agent inside what researchers call the \"inner harness\" — the model's native reasoning loop, tool-calling manifest, and unstructured context window — without wrapping it in the \"outer harness\" that would have enforced read-only defaults, required cryptographic human sign-off on material writes, and recorded the decision so it could be replayed.",[190,197,198],{},"That is not a model failure. It is an architecture failure.",[190,200,201],{},"Frontier large language models score well on isolated benchmarks. They achieve over 70 percent success rates on synthetic code-generation tasks like SWE-Bench Verified. But when confronted with long-horizon execution across evolving enterprise systems — where \"the environment\" is the company's live financial stack, not a stable code repository — task resolution rates collapse to under 26 percent on SWE-Bench Pro and 21 percent on SWE-EVO. The model's reasoning is intact. What fails is the operational envelope: which systems the model may touch, which changes require approval, and whether the decision is recorded in a form that survives the chat session.",[190,203,204],{},"This essay describes the outer harness — the deterministic control layer enterprises must build around autonomous agents — and the bi-temporal lifecycle graph that makes AI-mediated decisions auditable, so the question \"who approved this\" has a documentary answer, not a log file you cannot parse.",[206,207,209],"h2",{"id":208},"inner-harness-versus-outer-harness-where-control-lives","Inner harness versus outer harness: where control lives",[190,211,212],{},"Most agent frameworks stop at the inner harness. The inner harness is what the model vendor provides: the reasoning loop that inspects an input, picks a tool from a list, calls that tool, reads the result, and repeats until the model decides it is done. That loop is where the raw capability lives — the pattern-matching, the next-token prediction, the fluency. It is also where enterprises have the least control.",[190,214,215],{},"The outer harness is what you build around that loop. It is the deterministic control layer that decides which tools the agent may call, which outputs require human approval before they commit, and what gets recorded so you can reconstruct the decision later. The inner harness is probabilistic. The outer harness is not. It enforces rules the model cannot override: read-only by default, writes blocked until a named person signs, and every proposal logged whether it was approved or refused.",[190,217,218],{},"When enterprises deploy agents without an outer harness — letting the model's inner loop talk directly to CRM, ERP, or financial systems — the failure mode is predictable. The agent does what it was instructed, but nobody enforced the constraint that should have been architectural, not conversational. \"Ask before you write\" is a conversational instruction. \"Writes are blocked at the API layer until a cryptographic approval token is present\" is an architectural constraint. Only the second one survives when the model misunderstands, or when an attacker tries prompt injection.",[190,220,221],{},"The outer harness has three layers:",[223,224,226],"h3",{"id":225},"feedforward-guides-shape-the-environment-before-the-model-reasons","Feedforward guides: shape the environment before the model reasons",[190,228,229],{},"Feedforward controls prevent errors before the model generates a single token. They work by constraining what the agent can see and what tools it can call, based on the job it is working on and the role of the person who invoked it.",[190,231,232,236],{},[233,234,235],"strong",{},"Workspace schema."," The structured definition of what this job requires: which data sources are in scope, which output format is expected, which approval rules apply. The schema is deterministic. It does not depend on the model interpreting a prompt correctly. It is the environment the model inherits when the session starts.",[190,238,239,242],{},[233,240,241],{},"Standard operating procedures (SOPs)."," The authoritative wiki, policy doc, or contract clause the agent must follow. Retrieved from the source of truth — not from a chat memory — and injected into context as immutable ground truth. If the model tries to cite a stale policy, the feedforward guide has already constrained the retrieval to return only active versions.",[190,244,245,248],{},[233,246,247],{},"Role-based tool permissions."," Not every agent session should have write access. The permissions the agent inherits depend on the role of the person who started the job. A finance analyst's session can read billing tables but cannot post to the general ledger. A controller's session can propose journal entries, but those entries still sit in escrow until the controller signs them. The model does not decide its own permissions. The outer harness does.",[223,250,252],{"id":251},"feedback-sensors-validate-outputs-before-they-commit","Feedback sensors: validate outputs before they commit",[190,254,255],{},"Feedback sensors run after the model produces an output but before that output is allowed to leave the enterprise or change a live system. They are deterministic checks, not subjective judgements.",[190,257,258,261],{},[233,259,260],{},"Abstract Syntax Tree (AST) parsers."," If the agent generated code, the AST parser verifies the code is syntactically valid before it can be deployed. If the agent generated a JSON payload for an API call, the schema validator checks it against the expected structure.",[190,263,264,267],{},[233,265,266],{},"Type checkers and linters."," Static analysis tools that catch errors the model might have introduced — undeclared variables, type mismatches, security anti-patterns. These run automatically, without human review, and block the output if they fail.",[190,269,270,273],{},[233,271,272],{},"Policy assertion checks."," If the agent proposed a discount, the policy check verifies the discount is within approved limits. If the agent proposed a contract clause, the policy check confirms the clause does not waive a right the legal team has declared non-negotiable. These are rules you can test in isolation. They do not depend on the model \"understanding\" the policy. They enforce it.",[190,275,276,279],{},[233,277,278],{},"Database constraint verifiers."," If the agent is proposing a write to a database, the constraint verifier checks foreign-key relationships, uniqueness constraints, and required fields before the write is allowed. A model that hallucinates a customer ID will be caught here, not after the write corrupts the production table.",[223,281,283],{"id":282},"the-steering-loop-convert-failures-into-permanent-fixes","The steering loop: convert failures into permanent fixes",[190,285,286],{},"When a feedback sensor catches an error, the outer harness does not simply tell the model \"try again.\" It routes the failure back as structured telemetry — which rule failed, which value was out of bounds — and, if the same class of error appears repeatedly, it converts the failure into a permanent guard: a tighter schema, a new pre-flight check, an additional approval gate.",[190,288,289,290,293],{},"The core rule of harness engineering, as systems architects have been teaching for years, is this: ",[233,291,292],{},"every operational failure must produce a structural change in the harness, not an ephemeral tweak to a conversational prompt."," If the agent keeps proposing journal entries that violate the company's capitalisation threshold, the fix is not \"please remember the threshold is $5,000.\" The fix is a constraint in the outer harness that automatically rejects any journal-entry payload where the debit exceeds $5,000 and no controller signature is attached.",[190,295,296],{},"That is the difference between hoping the model behaves and enforcing the behaviour architecturally.",[206,298,300],{"id":299},"mechanics-of-the-enterprise-write-gate-and-privileged-non-human-identities","Mechanics of the enterprise write-gate and privileged non-human identities",[190,302,303],{},"The critical threat surface of enterprise generative AI is not the generation of inaccurate text, but the unauthorized execution of state changes across systems of record. When a sub-agent transitions from read-only contextual retrieval to initiating database writes, executing automated clearing house (ACH) transfers, deploying compiled binaries, or transmitting external communications, it crosses from an informational tool to an entitled system user.",[190,305,306],{},"IBM’s 2025 Cost of a Data Breach report established that 13% of surveyed organizations experienced breaches directly involving an AI model or application. Within those compromised organizations, an overwhelming 97% lacked adequate AI access controls. The attack vectors leading to these incidents were dominated by supply chain vulnerabilities—specifically over-privileged Application Programming Interface (API) connectors, misconfigured application plugins, and unvalidated third-party tool bindings. The same research documented that unauthorized shadow AI featured in 20% of enterprise security incidents, elevating average data breach damages by approximately $670,000 due to extensive exfiltration of intellectual property and personally identifiable information (PII).",[190,308,309],{},"To eliminate this vulnerability, the outer harness must enforce an architectural invariant: read-only is the non-negotiable default state of the enterprise AI substrate. An assistant permitted to read records, inspect invoices, and draft reconciliation proposals cannot directly breach the financial integrity of a business. A model wired directly to a payments or customer database via a write-enabled service token represents an unmonitored privileged identity that operates without fatigue, without human hesitation, and without performance reviews. The OWASP Top 10 for Large Language Model Applications identifies indirect prompt injection (LLM01) and excessive agency (LLM06) as primary enterprise vulnerabilities. If an external, untrusted input—such as an inbound vendor invoice containing hidden instructions—can manipulate a sub-agent’s internal reasoning, an over-privileged connector will execute that instruction with the full authorization of the underlying service token.",[190,311,312],{},"Enforcing the write-gate requires implementing a rigorous non-human identity governance framework:",[314,315,316,323,329],"ol",{},[317,318,319,322],"li",{},[233,320,321],{},"Non-human user joiner-mover-leaver (JML) lifecycles."," Sub-agents must never run under shared administrative credentials or static developer tokens created to expedite a pilot. Every sub-agent instance operating within an active workstream must receive an ephemeral, cryptographically distinct identity tied to a named operational owner, bound by a deterministic expiration timestamp, and restricted to fine-grained scopes.",[317,324,325,328],{},[233,326,327],{},"Payload staging and multi-tier approval gates."," When a sub-agent issues a write-call, the outer harness intercepts the command at the network layer. The payload is placed into transactional escrow, and an execution gate evaluates the blast radius. Modifications below pre-defined risk and spend ceilings that satisfy deterministic schema validation may clear automatically, while actions exceeding financial or operational thresholds require explicit, cryptographically signed human authorization before the packet is dispatched to the enterprise system of record.",[317,330,331,334],{},[233,332,333],{},"Mandatory refusal logging."," If an operator denies a sub-agent's proposed state change, or if a feedback sensor flags an unauthorized tool invocation, the outer harness records the refusal payload, the associated context state, and the rejecting authority. A security and audit architecture that logs only successful API transactions is functionally blind; an absence of recorded refusals demonstrates to internal auditors that safety controls are bypassed rather than actively filtering risk.",[190,336,337,338,342],{},"The legal and financial necessity of this write-gate architecture is demonstrated in established administrative case law. In ",[339,340,341],"em",{},"Moffatt v. Air Canada"," (2024 BCCRT 149), the Civil Resolution Tribunal rejected the airline's defense that its automated customer-facing chatbot constituted an independent legal entity responsible for its own misstatements regarding bereavement fares. The tribunal ruled that an organization maintains absolute vicarious liability for the representations, automated concessions, and commitments made by its digital agents, regardless of whether the output was an unmonitored hallucination or an approved transmission. When a sub-agent's write action impacts a consumer or an official ledger, the enterprise cannot disclaim the outcome; it must produce the documentary lineage proving how the action was validated, approved, and released.",[344,345,346,365],"table",{},[347,348,349],"thead",{},[350,351,352,356,359,362],"tr",{},[353,354,355],"th",{},"Architectural dimension",[353,357,358],{},"Bare model \u002F consumer shadow AI",[353,360,361],{},"Inner-loop framework (e.g., raw LangChain\u002FAutoGen)",[353,363,364],{},"Enterprise outer harness (governed operating envelope)",[366,367,368,383,397,411,425,439],"tbody",{},[350,369,370,374,377,380],{},[371,372,373],"td",{},"Execution boundary",[371,375,376],{},"Unmanaged third-party multi-tenant cloud.",[371,378,379],{},"Local container or virtual machine; direct API dispatch.",[371,381,382],{},"Isolated, tenant-confined workspace; network-sandboxed.",[350,384,385,388,391,394],{},[371,386,387],{},"Identity and privilege",[371,389,390],{},"Anonymous user or personal consumer account.",[371,392,393],{},"Shared developer API token; broad admin rights.",[371,395,396],{},"Ephemeral non-human user credentials; strict JML lifecycle.",[350,398,399,402,405,408],{},[371,400,401],{},"Tool execution policy",[371,403,404],{},"Unrestricted natural-language browser output.",[371,406,407],{},"Unconstrained autonomous function-calling loops.",[371,409,410],{},"Read-only default; staged payload escrow on writes.",[350,412,413,416,419,422],{},[371,414,415],{},"Verification and evals",[371,417,418],{},"None; implicit trust in generated text.",[371,420,421],{},"Subjective model-as-judge prompt wrappers.",[371,423,424],{},"Deterministic AST parsers, linters, and schema verifiers.",[350,426,427,430,433,436],{},[371,428,429],{},"Audit and state lineage",[371,431,432],{},"Ephemeral browser session; unlogged.",[371,434,435],{},"Flat execution trace logs; transient memory.",[371,437,438],{},"Cryptographically signed, bi-temporal Lifecycle Graph.",[350,440,441,444,447,450],{},[371,442,443],{},"Failure recovery mode",[371,445,446],{},"Manual human re-prompting on terminal error.",[371,448,449],{},"Indefinite programmatic retry loops; token exhaustion.",[371,451,452],{},"Feedback sensor routing to steering loop; human intervention.",[206,454,456],{"id":455},"why-vector-search-breaks-on-enterprise-policies-the-temporal-blindness-problem","Why vector search breaks on enterprise policies: the temporal blindness problem",[190,458,459,460,465],{},"Enterprise AI suffers from a memory problem. Employees use disconnected tools — one chatbot for summarising, another for drafting, a third for financial analysis — and spend hours copying outputs between systems. That fragmentation creates the illusion of speed at the individual level while organisational throughput stalls. ",[461,462,464],"a",{"href":463},"https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Fworklab\u002Fwork-trend-index\u002Fai-at-work-is-here-now-comes-the-hard-part","Microsoft and LinkedIn's research"," found that 75 percent of knowledge workers use generative AI, and 78 percent bring their own tools. The official path is too slow or too restrictive, so people route work through personal accounts on phones.",[190,467,468,469,473],{},"Banning those tools does not solve the problem. After ",[461,470,472],{"href":471},"https:\u002F\u002Fwww.cnbc.com\u002F2023\u002F05\u002F02\u002Fsamsung-bans-use-of-ai-like-chatgpt-for-staff-after-misuse-of-chatbot.html","Samsung engineers leaked proprietary code to ChatGPT in 2023",", enterprises issued blanket bans. Employees routed around them. The organisation lost visibility into where intellectual property was going, but the leakage continued. A ban without a usable sanctioned substitute just moves the work underground.",[190,475,476,477],{},"To fix fragmentation, enterprise architecture teams deploy Retrieval-Augmented Generation (RAG) — systems that retrieve relevant documents from a vector database and inject them into the model's context. That works well for static corpora. It breaks on evolving enterprise policies because vector search has a structural flaw: ",[233,478,479],{},"temporal blindness.",[190,481,482],{},"Vector databases calculate semantic similarity — how close two pieces of text are in meaning — but they do not understand time. If your company has a 2022 travel policy, a 2024 draft revision, and a 2026 active mandate, the vector database sees all three as equally relevant if they use similar language. When a customer-service agent asks \"what is our cancellation policy,\" the retrieval engine might return the 2022 version because its phrasing happens to match the query slightly better than the 2026 text.",[190,484,485],{},"The model, given three conflicting policies, synthesizes an answer. That answer might promise a refund the 2026 policy does not allow. The company is liable. The customer has a screenshot. The vector database did not know which document was \"the truth.\"",[190,487,488,489,493],{},"Graph-based systems like ",[461,490,492],{"href":491},"https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Fresearch\u002Fproject\u002Fgraphrag\u002F","GraphRAG"," improve on flat vector search by clustering related documents and pre-generating summaries. But pre-summarized graphs still require batch re-indexing when policies change. If a policy changes at 9:00 AM, agents running at 9:01 AM need to see the new rule. Auditors investigating a 8:59 AM transaction need to see the old rule. Static graphs cannot answer both questions because they overwrite history when they update.",[190,495,496,497,500,501,504],{},"What enterprises need is a memory system that tracks ",[233,498,499],{},"what was true when",", not just ",[233,502,503],{},"what is true now",".",[206,506,508],{"id":507},"the-lifecycle-graph-memory-that-tracks-what-was-true-when","The lifecycle graph: memory that tracks what was true when",[190,510,511],{},"To solve the temporal blindness problem, enterprises need a memory system that never deletes history — it just marks when each fact stopped being true and when the next version started. That is bi-temporal modeling, a technique from database theory (Richard Snodgrass, ISO SQL:2011) that tracks two independent timelines:",[190,513,514,517],{},[233,515,516],{},"1. Valid time — when it was true in the real world."," A travel policy is valid from January 1, 2026 to December 31, 2026. A customer contract amendment is valid retroactive to September 1. Valid time answers the business question: \"what rule governed this transaction on this date?\"",[190,519,520,523],{},[233,521,522],{},"2. Transaction time — when the system learned about it."," The policy was entered into the database on December 15, 2025. The amendment was signed on September 28 but recorded on October 2. Transaction time answers the audit question: \"what did the system know when it made this decision?\"",[190,525,526],{},"With both timelines, you can answer two critical questions that vector search cannot:",[528,529,530,536],"ul",{},[317,531,532,535],{},[233,533,534],{},"Current state (for operations):"," What is the active policy right now? Query for records where valid time includes today and transaction time is still open.",[317,537,538,541],{},[233,539,540],{},"Point-in-time reconstruction (for audits):"," What policy was active on September 15, and what did the system know about it on that date? Query for records where valid time included September 15 and transaction time included September 15.",[190,543,544],{},"The lifecycle graph never overwrites. When a policy changes, the old version is closed (valid-time-end = policy change date, transaction-time-end = system commit timestamp) and a new version is appended with a link showing it supersedes the old one. Nine months later, auditors can replay the exact state of the world when the agent made a decision. That is the documentary answer regulators and insurers need.",[223,546,548],{"id":547},"the-three-layers-of-the-graph","The three layers of the graph",[190,550,551],{},"The lifecycle graph organizes memory in three tiers:",[190,553,554,557],{},[233,555,556],{},"Episodic memory:"," Every prompt, tool call, approval, refusal, and API response is logged with exact timestamps. This is the forensic layer. When someone asks \"who approved this $4.2M journal entry,\" the graph has the payload, the controller's cryptographic signature, and the timestamp.",[190,559,560,563],{},[233,561,562],{},"Semantic memory:"," Business entities, roles, projects, policies, and their relationships. Each edge carries valid-time intervals. When an agent asks \"what is the capitalisation threshold,\" it traverses only edges where valid-time includes today. Stale policies are in the graph but filtered out.",[190,565,566,569],{},[233,567,568],{},"Community memory:"," Higher-level patterns and interdependencies. \"Revenue recognition rules interact with contract amendment workflows, which are blocked during deployment freezes.\" This layer helps agents understand context without re-deriving it from raw events.",[190,571,572,573,576],{},"The key architectural rule: ",[233,574,575],{},"old facts are never deleted, only invalidated."," That discipline is what makes the system auditable. When a regulator asks what the agent saw on a given date, you can rewind the graph to that exact state. When an operator asks what the current rule is, the graph filters to active facts only.",[206,578,580],{"id":579},"operationalizing-multi-agent-workstreams-in-complex-enterprise-workflows","Operationalizing multi-agent workstreams in complex enterprise workflows",[190,582,583],{},"To evaluate how the Outer Harness and Lifecycle Graph operate in production, consider an enterprise executing a cross-functional financial close: the Q3 Financial Close and Revenue Recognition determination governed by ASC 606 under evolving customer contract amendments.",[190,585,586],{},"In an unmanaged environment, finance analysts paste contract excerpts into disparate chat windows, attempt manual spreadsheet reconciliations, and copy unverified journal adjustments into the general ledger. If an amendment was signed on September 28 but retroactive to September 1, a standard RAG system pulling contract files risks citing original payment terms rather than amended revenue milestones, leading to inaccurate revenue recognition and subsequent restatements.",[190,588,589],{},"Within the governed operating envelope, the entire workflow is managed as a formal workstream where specialized sub-agents and operational state transitions progress through strictly enforced lifecycle gates.",[344,591,592,611],{},[347,593,594],{},[350,595,596,599,602,605,608],{},[353,597,598],{},"Stage",[353,600,601],{},"Trigger \u002F input",[353,603,604],{},"Architectural mechanism",[353,606,607],{},"Temporal state and graph operation",[353,609,610],{},"Resulting state and governance gate",[366,612,613,630,647,664,681,698,715],{},[350,614,615,618,621,624,627],{},[371,616,617],{},"1. Workstream initiation",[371,619,620],{},"Controller executes charter for Q3 close.",[371,622,623],{},"Outer Harness workspace orchestrator generates an ephemeral execution sandbox.",[371,625,626],{},"Instantiates a unique workstream node on the episodic subgraph; records transaction start time.",[371,628,629],{},"Ephemeral non-human user token issued; permissions restricted to read-only financial data.",[350,631,632,635,638,641,644],{},[371,633,634],{},"2. Context grounding",[371,636,637],{},"Workspace orchestrator initiates retrieval.",[371,639,640],{},"Enterprise wiki and semantic graph engine query active accounting standards.",[371,642,643],{},"Traverses the semantic graph for active accounting SOPs where valid time contains the close date.",[371,645,646],{},"Filters out deprecated 2024 revenue guidance; returns only active ASC 606 corporate rules.",[350,648,649,652,655,658,661],{},[371,650,651],{},"3. Contract ingestion and triangulation",[371,653,654],{},"Grounded brief assigned to legal and finance sub-agents.",[371,656,657],{},"Legal and finance sub-agents read billing tables and contract amendments.",[371,659,660],{},"Executes bi-temporal edge traversals to evaluate contract amendments across validity windows.",[371,662,663],{},"Read-only connectors prevent modifications to CRM, billing records, or customer vaults.",[350,665,666,669,672,675,678],{},[371,667,668],{},"4. Sensor validation check",[371,670,671],{},"Finance sub-agent drafts a journal adjustment.",[371,673,674],{},"Outer Harness feedback sensors execute deterministic mathematical validation.",[371,676,677],{},"Evaluates proposed debits against credits; validates the cited amendment node against the signature ledger.",[371,679,680],{},"Sensor flag: unexecuted amendment cited; triggers the steering loop to search for the executed addendum.",[350,682,683,686,689,692,695],{},[371,684,685],{},"5. Staged write proposal",[371,687,688],{},"Sub-agent resolves the citation and prepares the payload.",[371,690,691],{},"Outer Harness transaction escrow intercepts the proposed general ledger entry ($1.4M).",[371,693,694],{},"Generates a staging node on the episodic subgraph capturing the proposed general ledger adjustment.",[371,696,697],{},"Write-gate locks execution: the proposed transaction exceeds the $100,000 automated ceiling.",[350,699,700,703,706,709,712],{},[371,701,702],{},"6. Human-in-the-loop release",[371,704,705],{},"Staging lock pages the corporate controller.",[371,707,708],{},"Governance console presents the diff, active policy versions, and citations on the Lifecycle Graph.",[371,710,711],{},"Graph presents the provenance chain linking ERP invoices, the executed addendum, and the SOP.",[371,713,714],{},"Controller signs the release token with a cryptographic hardware key; stores an immutable approval record.",[350,716,717,720,723,726,729],{},[371,718,719],{},"7. Ledger commit and state append",[371,721,722],{},"Validated cryptographic release token received.",[371,724,725],{},"Scoped write connector executes the transaction against the general ledger API.",[371,727,728],{},"Appends a new ledger-state entity; invalidates the prior balance edge; records valid time and transaction time.",[371,730,731],{},"A single atomic write is committed; ephemeral non-human agent credentials are automatically revoked.",[190,733,734],{},"This multi-agent workflow demonstrates compounding intelligence. The output of the revenue recognition determination does not vanish into a chat log. It is committed to the Lifecycle Graph as an immutable, interconnected structure. When internal auditors, tax compliance authorities, or FP&A teams conduct analyses in future quarters, they do not review an ungrounded textual summary. They traverse the graph to inspect the exact brief, the active policy version, the evidence retrieved, the sensor validation traces, the controller’s signature, and the resulting ledger transition.",[206,736,738],{"id":737},"regulatory-reconstruction-and-evidentiary-defensibility-under-active-enforcement","Regulatory reconstruction and evidentiary defensibility under active enforcement",[190,740,741],{},"The deployment of enterprise AI has entered an era of direct enforcement, where administrative agencies evaluate concrete operational systems rather than high-level ethical statements.",[190,743,744],{},"In March 2024, the U.S. Securities and Exchange Commission (SEC) announced settled charges against investment advisers Delphia (USA) Inc. and Global Predictions Inc., imposing $400,000 in total civil penalties for making false and misleading public statements regarding their deployment of artificial intelligence. The SEC’s orders established that advertising automated capabilities, algorithmically predictive models, or \"AI-run\" processes when internal operations rely on manual interventions or conventional automation violates basic statutory protections against misleading statements of material fact.",[190,746,747],{},"Similarly, the Federal Trade Commission (FTC) warned organizations that claims regarding algorithmic accuracy, automated oversight, and fairness must be substantiated by demonstrable operational evidence before publication. In the European Union, Regulation (EU) 2024\u002F1689 (the EU AI Act) establishes statutory documentation, data governance, and continuous human oversight duties for high-risk systems, mandating that operators possess the architectural capability to interrupt, override, and reconstruct automated decisions throughout their operational lifecycles.",[190,749,750],{},"Organizations cannot satisfy regulatory examiners or defense standards with narrative slide decks, vendor marketing claims, or steering committee minutes. Regulatory bodies, external financial auditors, and commercial insurers require a verifiable reconstruction pack for every material automated action:",[528,752,753,759,765,771,777],{},[317,754,755,758],{},[233,756,757],{},"The exact ingestion state."," The verbatim input payload, prompt context, and system instructions dispatched to the model runtime, preserved without post-hoc summarization.",[317,760,761,764],{},[233,762,763],{},"The active policy version."," The specific regulatory rule, corporate standard, or price sheet active at the moment of execution, verified via immutable bi-temporal valid-time intervals rather than current-state documentation.",[317,766,767,770],{},[233,768,769],{},"The entitled human sign-off."," The cryptographic identity of the human operator who reviewed the staged payload and approved the write-gate release, or the deterministic rules proving why an automated write cleared below a sanctioned risk ceiling.",[317,772,773,776],{},[233,774,775],{},"The sensor verification record."," The programmatic telemetry demonstrating that the sub-agent output successfully cleared deterministic AST parsers, schema validations, and security linters before commit.",[317,778,779,782],{},[233,780,781],{},"The comprehensive refusal log."," The historical ledger of actions the system actively blocked or rejected, proving that governance controls function as active physical barriers rather than passive policy advisories.",[190,784,785],{},"When enterprise leadership aligns systems architecture with ISO\u002FIEC 42001 (the international management system standard for artificial intelligence) or the NIST AI Risk Management Framework (AI RMF), the Outer Harness and Lifecycle Graph supply the required structural artifacts. Compliance ceases to be an annual, retrospective paper-gathering exercise. It functions as the continuous, mathematical byproduct of how digital work is executed, audited, and preserved across the enterprise.",[206,787,789],{"id":788},"a-call-to-ctos-and-enterprise-architects","A call to CTOs and enterprise architects",[190,791,792],{},"Scaling autonomous AI is an engineering and governance problem, not a model-evaluation contest. Frontier models have impressive reasoning capability. That capability is worthless if your enterprise cannot constrain where the model points, govern which systems it can change, and reconstruct how a decision was made nine months later when the auditor asks.",[190,794,795],{},"Organisations that let agents operate via unstructured inner loops — the model's native tool-calling manifest and conversational memory — without wrapping them in a deterministic outer harness will generate compliance liabilities, security breaches, and untraceable financial errors. The fix is architectural:",[190,797,798,801],{},[233,799,800],{},"1. Enforce read-only by default."," Writes are blocked at the API layer until a named human approves the exact payload. That approval is cryptographically signed and logged. Not \"the model asked first.\" The payload waits in escrow.",[190,803,804,807],{},[233,805,806],{},"2. Build feedforward guides and feedback sensors."," Shape the environment before the model reasons: which tools it can call, which policies are active, which role it inherits. Validate outputs before they commit: AST parsers, schema checkers, policy assertions. Convert repeated failures into permanent structural guards, not conversational reminders.",[190,809,810,813],{},[233,811,812],{},"3. Deploy a bi-temporal lifecycle graph."," Track what was true when (valid time) and when the system learned about it (transaction time). Never overwrite history; invalidate it. That is how you answer \"what policy governed this transaction\" (operations) and \"what did the system know when it decided\" (audit) without reconstructing chat logs.",[190,815,816],{},"The worked example in this essay — Q3 financial close, revenue recognition, contract amendments, staged writes, controller sign-off — is the operational reality of enterprise AI. The model did its job. The architecture either prevented a $4.2M error with no human review, or it allowed it. That difference is the outer harness.",[190,818,819],{},"Sustainable enterprise AI means every autonomous action occurs inside a deterministic envelope that enforces access, approval, and auditability. The inner harness is what the model vendor gives you. The outer harness is what you build to make it safe. Build it before the incident teaches you why it was necessary.",[821,822],"hr",{},[206,824,826],{"id":825},"references","References",[528,828,829,835,840,846,852,857,862,868,874,880,886,892],{},[317,830,831],{},[461,832,834],{"href":833},"https:\u002F\u002Fwww.cbc.ca\u002Fnews\u002Fcanada\u002Fbritish-columbia\u002Fair-canada-chatbot-lawsuit-1.7116416","CBC News, Air Canada found liable for chatbot’s bad advice on plane tickets (15 February 2024)",[317,836,837],{},[461,838,839],{"href":471},"CNBC, Samsung bans staff use of generative AI after misuse (2 May 2023)",[317,841,842],{},[461,843,845],{"href":844},"https:\u002F\u002Fnewsroom.ibm.com\u002F2025-07-30-ibm-report-13-of-organizations-reported-breaches-of-ai-models-or-applications,-97-of-which-reported-lacking-proper-ai-access-controls","IBM Security, Cost of a Data Breach Report 2025 (30 July 2025)",[317,847,848],{},[461,849,851],{"href":850},"https:\u002F\u002Fwww.iso.org\u002Fstandard\u002F81230.html","ISO\u002FIEC 42001:2023, Information technology — Artificial intelligence — Management system",[317,853,854],{},[461,855,856],{"href":463},"Microsoft and LinkedIn, 2024 Work Trend Index Annual Report (8 May 2024)",[317,858,859],{},[461,860,861],{"href":491},"Microsoft Research, GraphRAG: Unlocking LLM discovery on narrative private data (12 June 2024)",[317,863,864],{},[461,865,867],{"href":866},"https:\u002F\u002Fwww.nist.gov\u002Fitl\u002Fai-risk-management-framework","NIST, Artificial Intelligence Risk Management Framework (AI RMF 1.0)",[317,869,870],{},[461,871,873],{"href":872},"https:\u002F\u002Fgenai.owasp.org\u002Fllm-top-10\u002F","OWASP Foundation, OWASP Top 10 for Large Language Model Applications",[317,875,876],{},[461,877,879],{"href":878},"https:\u002F\u002Fwww.canlii.org\u002Fen\u002Fbc\u002Fbccrt\u002Fdoc\u002F2024\u002F2024bccrt149\u002F2024bccrt149.html","Civil Resolution Tribunal of British Columbia, Moffatt v. Air Canada, 2024 BCCRT 149",[317,881,882],{},[461,883,885],{"href":884},"https:\u002F\u002Feur-lex.europa.eu\u002Flegal-content\u002FEN\u002FTXT\u002F?uri=OJ:L_202401689","Regulation (EU) 2024\u002F1689, the EU AI Act",[317,887,888],{},[461,889,891],{"href":890},"https:\u002F\u002Fwww.ftc.gov\u002Fbusiness-guidance\u002Fblog\u002F2023\u002F02\u002Fkeep-your-ai-claims-check","U.S. Federal Trade Commission, Keep your AI claims in check (27 February 2023)",[317,893,894],{},[461,895,897],{"href":896},"https:\u002F\u002Fwww.sec.gov\u002Fnewsroom\u002Fpress-releases\u002F2024-36","U.S. Securities and Exchange Commission, Press Release 2024-36 (18 March 2024)",{"title":169,"searchDepth":170,"depth":170,"links":899},[900,906,907,908,911,912,913,914],{"id":208,"depth":170,"text":209,"children":901},[902,904,905],{"id":225,"depth":903,"text":226},3,{"id":251,"depth":903,"text":252},{"id":282,"depth":903,"text":283},{"id":299,"depth":170,"text":300},{"id":455,"depth":170,"text":456},{"id":507,"depth":170,"text":508,"children":909},[910],{"id":547,"depth":903,"text":548},{"id":579,"depth":170,"text":580},{"id":737,"depth":170,"text":738},{"id":788,"depth":170,"text":789},{"id":825,"depth":170,"text":826},"2026-09-22","In Q3 2024, a sub-agent proposed $4.2M in journal entries with no human review. The model was fine. The architecture was not. Here is how the outer harness prevents that.",{"eyebrow":918,"title":919},"Short answers","A deterministic envelope around the agents",[921,924,927],{"question":922,"answer":923},"Why do frontier models stall on enterprise work?","They pass isolated coding benchmarks but fail on long-horizon jobs where the company — not a code repo — is the environment, and where a wrong write costs money or reputation.",{"question":925,"answer":926},"What is the outer harness?","The deterministic envelope around autonomous agents: which systems they may read, which changes require approval, and what gets recorded so you can reconstruct the decision later.",{"question":928,"answer":929},"What is the lifecycle graph for?","A bi-temporal record of work — what was true when the decision was made, and when the system learned about it — so auditors and operators can replay the exact state that informed an AI-proposed change.","\u002Fblog\u002Fthe-architecture-of-the-outer-harness",{"title":179,"description":916},"insight","blog\u002Fthe-architecture-of-the-outer-harness",[935,936,937,938],"thought-leadership","agent-harness","architecture","governance","diAtArZHQfY63s_ZWJA55RQJM4A3Z2SwIjEhWUaXWHA",{"hero":941,"id":943,"title":944,"archived":163,"authors":164,"badge":164,"body":945,"date":164,"definedTerm":164,"department":164,"description":949,"extension":172,"eyebrow":950,"faqHeader":164,"faqs":164,"footerBand":951,"headline":164,"image":164,"industry":164,"jobType":164,"listed":131,"location":164,"navigation":131,"openRoles":164,"pageLayout":164,"path":60,"relatedHeading":957,"seo":958,"series":164,"sitemap":131,"status":164,"stem":959,"subhead":164,"tags":164,"video":164,"whyJoin":164,"workplaceType":164,"__hash__":960},{"filename":942},"u2221455217_Flat_design_of_a_futuristic_minimalist_landscape__5d589295-cdea-4ea9-a262-be766881accf_1.png","content\u002Fblog\u002Findex.md","Exploring the future of intelligence.",{"type":166,"value":946,"toc":947},[],{"title":169,"searchDepth":170,"depth":170,"links":948},[],"Deep dives into pre-cognitive intelligence, sentient enterprises, and the evolving landscape of AI-driven business transformation.","Latest Research",{"headline":952,"description":953,"primaryLabel":954,"primaryTo":955,"secondaryLabel":956,"secondaryTo":12},"Stay at the frontier.","Subscribe for product updates and new insights.","Subscribe","\u002Fnewsletter","Explore the platform","More research",{"title":944,"description":949},"blog\u002Findex","BFSWGYO9bcTlaulivKYWyg08_DJHsdGg3OC6g_CG1Hw",[962,1246],{"id":963,"title":964,"archived":163,"authors":965,"badge":968,"body":970,"date":915,"definedTerm":980,"department":164,"description":1221,"extension":172,"eyebrow":164,"faqHeader":1222,"faqs":1224,"footerBand":164,"headline":164,"image":164,"industry":164,"jobType":164,"listed":131,"location":164,"navigation":131,"openRoles":164,"pageLayout":164,"path":1237,"relatedHeading":164,"seo":1238,"series":1239,"sitemap":131,"status":164,"stem":1240,"subhead":164,"tags":1241,"video":164,"whyJoin":164,"workplaceType":164,"__hash__":1245},"content\u002Fblog\u002Fthe-shared-canvas-for-human-ai-teams.md","How to Design Collaborative Canvas UX for Human-AI Teams",[966],{"name":967,"to":184},"Nimbus Research",{"label":969},"Explainer",{"type":166,"value":971,"toc":1209},[972,975,982,986,989,992,995,999,1002,1006,1013,1017,1023,1027,1033,1037,1040,1044,1047,1137,1141,1144,1155,1169,1173,1176,1196],[190,973,974],{},"Designing effective user experiences for artificial intelligence applications has historically meant refining simple text-based conversational interfaces. However, as enterprise software shifts from single-user prompts to multi-user collaborative execution, the traditional chat window rapidly reveals its cognitive and structural limitations. Complex team initiatives — such as product roadmap design, systems architecture mapping, and strategic campaign planning — are inherently spatial, nonlinear, and multi-dimensional. Forcing these collaborative processes into a single-column, chronological text feed limits human creativity and creates visual clutter. Designing modern collaborative canvas interfaces for human and artificial intelligence teams requires a complete evolution of user interface design patterns, shifting focus toward spatial layouts, real-time visual presence indicators, non-destructive editing overlays, and context-aware selection mechanisms that maintain human agency while harnessing background machine capabilities.",[190,976,977,978,981],{},"A ",[233,979,980],{},"collaborative AI canvas"," is a spatial user interface designed for real-time multi-user interaction, where human cursors and autonomous AI agents co-exist, edit, and manipulate visual nodes simultaneously. Unlike linear text boxes, spatial canvas interfaces utilize multi-user cursors, non-destructive draft overlays, context selection, and real-time diffing to enable clear attribution between human edits and machine-generated content.",[206,983,985],{"id":984},"beyond-the-chatbox-the-case-for-spatial-ai-collaboration","Beyond the chatbox: the case for spatial AI collaboration",[190,987,988],{},"Linear messaging and chat windows force complex, multi-dimensional team thoughts into a single chronological stream. When cross-functional teams attempt to collaborate inside a traditional artificial intelligence chat box, contextual nuance is rapidly lost, multi-user edits become fragmented across long threads, and long-form strategic planning becomes almost impossible to navigate efficiently.",[190,990,991],{},"Spatial canvas interfaces solve this structural limitation by arranging information across two dimensions. Adding autonomous artificial intelligence agents directly into these spatial environments creates a dynamic workspace where human professionals and digital models co-create without blocking each other's view, focus zones, or operational workflows.",[190,993,994],{},"By expanding the interaction surface into a spatial canvas, organizations allow team members to see the full structural context of a project at a glance, zoom into specific operational components, and observe background artificial intelligence agents generating content in adjacent nodes without disrupting active human typing.",[206,996,998],{"id":997},"core-interaction-patterns-for-human-ai-canvas-ux","Core interaction patterns for human-AI canvas UX",[190,1000,1001],{},"Designing effective multiplayer artificial intelligence canvases requires explicit visual interaction patterns to manage human cognitive load, maintain visual clarity, and protect user agency.",[223,1003,1005],{"id":1004},"ghost-cursors-and-ambient-ai-presence","Ghost cursors and ambient AI presence",[190,1007,1008,1009,1012],{},"Human users rely heavily on spatial awareness to avoid editing the exact same text line or design box simultaneously. In multiplayer artificial intelligence applications, active digital agents display distinct visual presences, such as ",[233,1010,1011],{},"ghost cursors",", animated bounding boxes, or pulsing rings, explicitly indicating the specific visual node or document section currently being processed by a language model. The cursor tells the room where the agent is working. It does not give the agent authority. A person still accepts or rejects the change.",[223,1014,1016],{"id":1015},"context-selection-rings","Context selection rings",[190,1018,1019,1020,1022],{},"When a human user prompts an artificial intelligence agent on a spatial canvas, they must explicitly define the source context required for execution. ",[233,1021,1016],{}," allow users to drag spatial connections or highlight specific visual canvas nodes, visibly linking source data — such as raw customer interview transcripts — directly to the generated output node, such as synthesized product requirements. On a company job, that selection still has to respect what the workspace was allowed to open. Pointing at a node is not a grant to every system the user could theoretically reach.",[223,1024,1026],{"id":1025},"non-destructive-draft-overlays","Non-destructive draft overlays",[190,1028,1029,1030,1032],{},"Artificial intelligence agents should never permanently overwrite human-generated canvas content without explicit human review. ",[233,1031,1026],{}," render machine suggestions in a highlighted or draft state, complete with clear inline controls for accepting, modifying, or rejecting proposed edits.",[223,1034,1036],{"id":1035},"spatial-auto-clustering","Spatial auto-clustering",[190,1038,1039],{},"When human teams perform rapid visual brainstorming by generating dozens of digital sticky notes, artificial intelligence agents can execute semantic clustering. The system groups notes by topic, generating concise category headers without disrupting active human placement.",[206,1041,1043],{"id":1042},"ui-component-taxonomy-for-collaborative-canvases","UI component taxonomy for collaborative canvases",[190,1045,1046],{},"The following matrix outlines standard user interface components for enterprise multiplayer human and artificial intelligence canvas applications.",[344,1048,1049,1065],{},[347,1050,1051],{},[350,1052,1053,1056,1059,1062],{},[353,1054,1055],{},"UI component",[353,1057,1058],{},"Human interface behavior",[353,1060,1061],{},"AI agent interface behavior",[353,1063,1064],{},"Conflict resolution and safeguard",[366,1066,1067,1081,1095,1109,1123],{},[350,1068,1069,1072,1075,1078],{},[371,1070,1071],{},"Multiplayer cursor",[371,1073,1074],{},"High-frequency pointer tracking",[371,1076,1077],{},"Focus indicator on the target node",[371,1079,1080],{},"Visual separation so cursors do not pretend to be the same actor",[350,1082,1083,1086,1089,1092],{},[371,1084,1085],{},"Selection highlight",[371,1087,1088],{},"Click-and-drag bounding box",[371,1090,1091],{},"Context attached to the selected nodes",[371,1093,1094],{},"Soft-locking on an active node during execution",[350,1096,1097,1100,1103,1106],{},[371,1098,1099],{},"Inline annotations",[371,1101,1102],{},"Manual comments and user tags",[371,1104,1105],{},"Automated validation and risk badges",[371,1107,1108],{},"Non-destructive draft overlay",[350,1110,1111,1114,1117,1120],{},[371,1112,1113],{},"Spatial nodes",[371,1115,1116],{},"Manual sticky note or shape creation",[371,1118,1119],{},"Grouping related notes",[371,1121,1122],{},"Versioned undo and redo",[350,1124,1125,1128,1131,1134],{},[371,1126,1127],{},"Execution controls",[371,1129,1130],{},"Direct click or keyboard shortcut",[371,1132,1133],{},"Loading indicator and stream progress",[371,1135,1136],{},"A stop the person on the job can trigger",[206,1138,1140],{"id":1139},"solving-spatial-collision-and-cognitive-overload","Solving spatial collision and cognitive overload",[190,1142,1143],{},"A major user experience risk in multiplayer artificial intelligence canvas design is visual chaos and cognitive overload. If four digital agents simultaneously populate dozens of spatial nodes while three human team members are actively typing, the workspace rapidly becomes unreadable and overwhelming.",[190,1145,1146,1147,1150,1151,1154],{},"To prevent cognitive overload and maintain operational focus, interface designers implement ",[233,1148,1149],{},"staging areas"," and ",[233,1152,1153],{},"focus isolation zones",":",[314,1156,1157,1163],{},[317,1158,1159,1162],{},[233,1160,1161],{},"The staging drawer."," Artificial intelligence agent generation occurs inside a collapsible side drawer or a minimized canvas node. The generated output is pushed to the primary visual canvas only after a human user triggers layout placement. The same idea applies to a live system: the draft can be ready, and the write still waits for a person.",[317,1164,1165,1168],{},[233,1166,1167],{},"Focus isolation zones."," When a human user enters a focused editing state on a specific canvas node, ambient background artificial intelligence generation in adjacent nodes is dimmed or temporarily paused, protecting the user's immediate cognitive focus.",[206,1170,1172],{"id":1171},"architectural-principles-for-canvas-state-binding","Architectural principles for canvas state binding",[190,1174,1175],{},"To bind a shared visual canvas node to both human user inputs and streaming artificial intelligence responses without breaking state consistency, system designers follow three foundational rules:",[528,1177,1178,1184,1190],{},[317,1179,1180,1183],{},[233,1181,1182],{},"Single source of truth."," All node coordinates, text content, and processing statuses are stored in a centralized shared state object rather than scattered across local component states.",[317,1185,1186,1189],{},[233,1187,1188],{},"Transactional updates."," When an artificial intelligence agent generates new text or creates visual shapes, each insertion or layout update is dispatched as a delta to the shared state, so connected human clients see a smooth render.",[317,1191,1192,1195],{},[233,1193,1194],{},"Explicit user ownership metadata."," Every node maintains clear metadata identifying whether the last modification was executed by a specific human team member or a designated artificial intelligence agent persona.",[190,1197,1198,1199,1203,1204,1208],{},"Nimbus holds this surface on a ",[461,1200,1202],{"href":1201},"\u002Fproduct\u002Fworkstreams\u002F","workstream",": the brief, the draft, and the refusal in one place. ",[461,1205,1207],{"href":1206},"\u002Fblog\u002Fsolo-ai-vs-multiplayer-ai\u002F","The shift from solo AI to multiplayer AI"," is why the private column stops being enough.",{"title":169,"searchDepth":170,"depth":170,"links":1210},[1211,1212,1218,1219,1220],{"id":984,"depth":170,"text":985},{"id":997,"depth":170,"text":998,"children":1213},[1214,1215,1216,1217],{"id":1004,"depth":903,"text":1005},{"id":1015,"depth":903,"text":1016},{"id":1025,"depth":903,"text":1026},{"id":1035,"depth":903,"text":1036},{"id":1042,"depth":170,"text":1043},{"id":1139,"depth":170,"text":1140},{"id":1171,"depth":170,"text":1172},"Visual design patterns, spatial interaction models, and human-in-the-loop review mechanisms for real-time multiplayer AI workspaces.",{"eyebrow":918,"title":1223},"A canvas, not a column",[1225,1228,1231,1234],{"question":1226,"answer":1227},"Why is a spatial canvas better than a chat window for team AI collaboration?","A spatial canvas allows information to be organized visually across two dimensions. Teams can break complex projects into parallel nodes, allowing different human users and AI agents to work on separate sections simultaneously without cluttering a single text feed.",{"question":1229,"answer":1230},"How do you prevent AI agents from overwriting human work on a canvas?","Collaborative canvases use non-destructive draft overlays and soft-locking mechanisms. AI-generated changes are presented as suggestions or draft layers that require human review and confirmation before they are committed.",{"question":1232,"answer":1233},"What are ghost cursors in multiplayer AI design?","Ghost cursors are visual indicators that display the active position and processing state of an AI agent on a shared canvas. They mirror human multiplayer cursors, helping team members track where an AI model is currently reading or writing. The cursor shows presence. It does not grant authority to release a change.",{"question":1235,"answer":1236},"How do canvas interfaces maintain performance with multiple streaming AI agents?","Performance is maintained using optimized client-side canvas rendering combined with efficient data sync protocols. Only transformed node coordinates and text updates are transmitted across the network, keeping interface frame rates smooth.","\u002Fblog\u002Fthe-shared-canvas-for-human-ai-teams",{"title":964,"description":1221},"explainer","blog\u002Fthe-shared-canvas-for-human-ai-teams",[1239,1242,1243,1244],"multiplayer AI","workstreams","collaborative AI","x9XdVHadKHtSR8bDd9_ZKAKNQtgNfMM0APUpBOAnQec",{"id":1247,"title":1248,"archived":163,"authors":1249,"badge":1251,"body":1252,"date":915,"definedTerm":1242,"department":164,"description":1733,"extension":172,"eyebrow":164,"faqHeader":1734,"faqs":1736,"footerBand":164,"headline":164,"image":164,"industry":164,"jobType":164,"listed":131,"location":164,"navigation":131,"openRoles":164,"pageLayout":164,"path":1749,"relatedHeading":164,"seo":1750,"series":1239,"sitemap":131,"status":164,"stem":1751,"subhead":164,"tags":1752,"video":164,"whyJoin":164,"workplaceType":164,"__hash__":1753},"content\u002Fblog\u002Fsolo-ai-vs-multiplayer-ai.md","Solo AI versus Multiplayer AI",[1250],{"name":182,"role":183,"to":184},{"label":969},{"type":166,"value":1253,"toc":1724},[1254,1257,1260,1266,1327,1331,1334,1337,1340,1347,1353,1357,1360,1366,1372,1378,1389,1439,1443,1446,1449,1455,1461,1467,1474,1556,1560,1563,1566,1569,1572,1575,1579,1582,1585,1588,1591,1594,1665,1669,1672,1675,1678,1684,1698,1700,1702],[190,1255,1256],{},"The pricing exception leaves on a Thursday. The owner is on leave. The assistant still has the thread. By Friday the CRM has a number the pricing desk never signed. The customer has it too.",[190,1258,1259],{},"That is solo AI doing what it was built to do. One person, one window, one fluent answer, one paste. The organisation then discovers that a personal tool does not become team software because the output was useful.",[190,1261,1262,1265],{},[233,1263,1264],{},"Multiplayer AI"," is not a louder chat. It is people and a model on the same job at the same time: the same brief, the same files, and a named person who can refuse a change. The unit is the job, not the seat and not the model. Two departments can open the work, see the same draft, and leave a record when someone says no.",[344,1267,1268,1281],{},[347,1269,1270],{},[350,1271,1272,1275,1278],{},[353,1273,1274],{},"What the programme was sold",[353,1276,1277],{},"What is true",[353,1279,1280],{},"What is still missing",[366,1282,1283,1294,1305,1316],{},[350,1284,1285,1288,1291],{},[371,1286,1287],{},"Everyone has AI",[371,1289,1290],{},"Everyone has a private window",[371,1292,1293],{},"A job more than one person can open",[350,1295,1296,1299,1302],{},[371,1297,1298],{},"Work is faster",[371,1300,1301],{},"First drafts are faster",[371,1303,1304],{},"A second person who can continue",[350,1306,1307,1310,1313],{},[371,1308,1309],{},"The team is aligned",[371,1311,1312],{},"The paste arrived",[371,1314,1315],{},"A refusal the desk can see",[350,1317,1318,1321,1324],{},[371,1319,1320],{},"The tool is official",[371,1322,1323],{},"Most use is still personal",[371,1325,1326],{},"A room the unofficial path does not replace",[206,1328,1330],{"id":1329},"solo-ai-is-a-good-personal-tool","Solo AI is a good personal tool",[190,1332,1333],{},"The first wave earned its keep. A quiet analyst drafts faster. A manager cleans a note. A specialist checks a clause before a meeting. None of that is a failure. It is a personal assistant doing personal work.",[190,1335,1336],{},"The failure starts when the personal window becomes the place the company job lives.",[190,1338,1339],{},"In a solo setup the employee prompts in isolation. The output — a price, a customer sentence, a forecast line — is copied into Slack, a document, or a system of record. The rest of the team meets the result without the prompt, without the files, and without a chance to refuse it before it travels. Information asymmetry is not a side effect. It is the architecture. Copy-paste is the integration. Provenance dies at the clipboard.",[190,1341,1342,1346],{},[461,1343,1345],{"href":1344},"https:\u002F\u002Fwww.mckinsey.com\u002Fcapabilities\u002Fquantumblack\u002Four-insights\u002Fthe-state-of-ai","McKinsey’s State of AI survey"," found that 88 percent of organisations use AI in at least one function, up from 78 percent a year earlier, while nearly two-thirds remain in experimentation or piloting and only about a third have begun to scale. A personal copilot is the experiment. Scale is the pricing desk opening the same exception on Friday and seeing the same brief, the same file, and the same unsigned number — still unsigned.",[190,1348,1349,1352],{},[461,1350,1351],{"href":463},"Microsoft and LinkedIn’s Work Trend Index"," makes the unofficial path explicit. Seventy-five percent of knowledge workers use generative AI at work and 78 percent of those users bring their own tools. Seventy-nine percent of leaders say their company needs AI to stay competitive, while 60 percent worry that leadership lacks a plan. Solo AI is already the default. The official licence is often a lower bound.",[206,1354,1356],{"id":1355},"why-the-private-chat-fails-the-team","Why the private chat fails the team",[190,1358,1359],{},"Three failures show up whenever a company job is done in a solo window.",[190,1361,1362,1365],{},[233,1363,1364],{},"Nobody else can continue."," The owner goes on leave. The thread stays in a personal account. The next person starts again, or guesses, or ships the last paste. Continuity was never a feature of the tool. Continuity was a person remaining at their desk.",[190,1367,1368,1371],{},[233,1369,1370],{},"Nobody else can check the inputs."," Surrounding teammates cannot see the original instruction, the files that were shown, or the step that was skipped. They can only see the fluent result. Review becomes taste. Taste is a poor control on a price.",[190,1373,1374,1377],{},[233,1375,1376],{},"Nobody else can refuse in time."," A solo chat has one user. The user who wants the meeting to end is the user who accepts. The person who would have stopped the number is in another function, on another login, looking at another window. By the time they see the CRM line, the customer has it.",[190,1379,1380,1384,1385,1388],{},[461,1381,1383],{"href":1382},"https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41562-021-01196-4","Yang and colleagues"," in ",[339,1386,1387],{},"Nature Human Behaviour"," (2022) found that remote work made collaboration networks more siloed, with fewer bridges between groups. Solo AI is a silo that feels like progress. Pricing, commercial, and the person who will sign are already three groups. The exception is one job. A personal chat makes the bridge a favour.",[344,1390,1391,1404],{},[347,1392,1393],{},[350,1394,1395,1398,1401],{},[353,1396,1397],{},"Solo failure",[353,1399,1400],{},"How it looks on Friday",[353,1402,1403],{},"What multiplayer has to hold",[366,1405,1406,1417,1428],{},[350,1407,1408,1411,1414],{},[371,1409,1410],{},"Nobody can continue",[371,1412,1413],{},"The owner is out; the thread is personal",[371,1415,1416],{},"The brief and the files on the job",[350,1418,1419,1422,1425],{},[371,1420,1421],{},"Nobody can check the inputs",[371,1423,1424],{},"A fluent number with no source list",[371,1426,1427],{},"What the model was shown",[350,1429,1430,1433,1436],{},[371,1431,1432],{},"Nobody can refuse in time",[371,1434,1435],{},"The CRM already moved",[371,1437,1438],{},"A named stop before the write",[206,1440,1442],{"id":1441},"what-multiplayer-actually-means","What multiplayer actually means",[190,1444,1445],{},"Multiplayer is easy to fake. Extra seats on a personal product is not it. A shared login is not it. A swarm of models handing tickets to each other is not it if no second human can open the job.",[190,1447,1448],{},"Multiplayer means the state is the job.",[190,1450,1451,1454],{},[233,1452,1453],{},"The same brief."," One instruction both departments can open. Not “as I told my assistant.” If two people cannot point at the same paragraph, they are still in solo mode with a meeting on top.",[190,1456,1457,1460],{},[233,1458,1459],{},"The same files."," The objects this run may read, on the job, as a list. A model that inherits “whatever the user pasted” is still a personal tool. When finance joins, they should see the same price file the model saw — not a reconstruction from chat exports.",[190,1462,1463,1466],{},[233,1464,1465],{},"The same refusal."," A visible no, with a name, on a class of change. Sync that keeps the document coherent is useful. It is not permission to change a live system. An agent inherits the privileges of the person who invoked it, or the permissions of the workspace it is operating in. It does not receive a broader grant than that person or that room. Read-only is the default until a person releases a write.",[190,1468,1469,1473],{},[461,1470,1472],{"href":1471},"https:\u002F\u002Fwww.melconway.com\u002FHome\u002FCommittees_Paper.html","Melvin Conway’s 1968 paper"," noted that organisations design systems that copy their communication structure. If the structure is one person and one chat, the AI copies that — including the bus factor of one. Multiplayer is the decision to copy the desk you already needed: the people who must stand on the result, in one room, with one stop.",[344,1475,1476,1488],{},[347,1477,1478],{},[350,1479,1480,1483,1486],{},[353,1481,1482],{},"Architectural feature",[353,1484,1485],{},"Solo AI",[353,1487,1264],{},[366,1489,1490,1501,1512,1523,1534,1545],{},[350,1491,1492,1495,1498],{},[371,1493,1494],{},"Who is in the session",[371,1496,1497],{},"One person, one model",[371,1499,1500],{},"Several people, and a model, on one job",[350,1502,1503,1506,1509],{},[371,1504,1505],{},"Where the brief lives",[371,1507,1508],{},"A prompt history",[371,1510,1511],{},"A paragraph the roster can open",[350,1513,1514,1517,1520],{},[371,1515,1516],{},"Where the files live",[371,1518,1519],{},"A paste, a download, a memory",[371,1521,1522],{},"Named objects on the job",[350,1524,1525,1528,1531],{},[371,1526,1527],{},"Where the refusal lives",[371,1529,1530],{},"The user closing the tab",[371,1532,1533],{},"A stored no on a class of write",[350,1535,1536,1539,1542],{},[371,1537,1538],{},"What the next person does",[371,1540,1541],{},"Starts again",[371,1543,1544],{},"Continues",[350,1546,1547,1550,1553],{},[371,1548,1549],{},"What a write requires",[371,1551,1552],{},"A paste",[371,1554,1555],{},"A named release",[206,1557,1559],{"id":1558},"speed-of-the-individual-speed-of-the-team","Speed of the individual, speed of the team",[190,1561,1562],{},"Solo AI is fast at the first draft. That is not the metric that fails. The metric that fails is time-to-alignment: how long it takes for a second department to stand on the same result.",[190,1564,1565],{},"In the solo pattern, alignment is a meeting. Draft, paste, comment, rewrite, paste again, book a slot, consolidate. Days are normal. The model made Tuesday faster and left Thursday untouched.",[190,1567,1568],{},"In a multiplayer pattern the alignment is the room. The brief is already there. The files are already there. The draft is a draft. The person who can refuse is on the roster, looking at the payload, not at a summary of a chat they were not in. Humans steer and sign. They do not reconstruct.",[190,1570,1571],{},"McKinsey’s survey is blunt about the distance between use and scale. Most organisations have not embedded AI deeply enough into workflows to realise material enterprise-level benefits. A workflow that still ends in a paste is not embedded. It is solo work with a company invoice.",[190,1573,1574],{},"You do not need multiplayer to summarise your own notes. You do need it when the output can change a CRM, a journal, or a sentence a customer will keep. You need several people when more than one owner must stand on the result, or when a handover will happen, or when a customer-facing line can leave. You need several models only when the hand-off already exists between human roles and you want a narrower tool for each step. Extra models without a shared job are still solo AI with a longer bill.",[206,1576,1578],{"id":1577},"questions-for-a-serious-distinction-review","Questions for a serious distinction review",[190,1580,1581],{},"A useful review does not start with the vendor’s topology diagram. It starts with last week’s exception.",[190,1583,1584],{},"If you remove every extra model, can two departments still share the files and the stop? If not, you never had multiplayer. You had a personal tool with more inference.",[190,1586,1587],{},"If you remove the second human, does the run still complete in private? If yes, you still have solo AI. The second seat was decoration.",[190,1589,1590],{},"Can you name the person who would have refused Friday’s CRM line, and show that they were on the job when the line was proposed? A node labelled “review” is not a name. A name who was not in the room is not a control.",[190,1592,1593],{},"Then the substitution question. Did the shared room replace the reconstruction meeting, or did it sit beside the personal chats and add a place to file the paste? Leaders are right to be suspicious of “alignment” that still begins with “which number did we send.”",[344,1595,1596,1609],{},[347,1597,1598],{},[350,1599,1600,1603,1606],{},[353,1601,1602],{},"Ask this",[353,1604,1605],{},"Multiplayer",[353,1607,1608],{},"Still solo",[366,1610,1611,1622,1633,1644,1655],{},[350,1612,1613,1616,1619],{},[371,1614,1615],{},"Can two departments open the same brief?",[371,1617,1618],{},"Yes, on the job",[371,1620,1621],{},"Only if someone forwards a thread",[350,1623,1624,1627,1630],{},[371,1625,1626],{},"Which files did the run read?",[371,1628,1629],{},"A named list",[371,1631,1632],{},"A recollection",[350,1634,1635,1638,1641],{},[371,1636,1637],{},"Who refuses a write?",[371,1639,1640],{},"A person on the roster",[371,1642,1643],{},"The user who wanted the meeting to end",[350,1645,1646,1649,1652],{},[371,1647,1648],{},"What happens when the owner is out?",[371,1650,1651],{},"The next person continues",[371,1653,1654],{},"The thread is gone",[350,1656,1657,1660,1663],{},[371,1658,1659],{},"Did a write wait?",[371,1661,1662],{},"A stored release, or a stored no",[371,1664,1552],{},[206,1666,1668],{"id":1667},"what-to-change-on-the-next-exception","What to change on the next exception",[190,1670,1671],{},"Pick one job that already crosses a desk. Name the owner. Put the brief on the job. Attach the files the model may read. Name the person who must refuse a change to a live system. Do not enable the write until that name has been in the room for a week and has recorded a no.",[190,1673,1674],{},"Separate the sanctioned path from the personal one. A ban that people route around on their phones is a policy, not a control. The official room has to be good enough that the personal subscription is no longer the path of least resistance. The Work Trend Index is the evidence: most people who use AI at work already bring their own. The unofficial path will keep pricing exceptions until the official one holds the brief, the files, and the stop.",[190,1676,1677],{},"Report the programme in the language of jobs, not seats. A director can challenge an exception. A director cannot challenge “adoption.” A month of spend and no refusals is a month you bought fluency without a teammate.",[190,1679,1680,1681,1683],{},"Nimbus is a Collaborative AI operating system. The room is a ",[461,1682,1202],{"href":1201},". The surface is a canvas. Reads are the default; a write waits for a person. The Lifecycle Graph keeps signed-off work as memory the organisation owns. It is not the ledger.",[190,1685,1686,1689,1690,1689,1693,1697],{},[461,1687,8],{"href":1688},"\u002Fcheckout",". ",[461,1691,1692],{"href":85},"Talk to sales",[461,1694,1696],{"href":1695},"\u002Fblog\u002Fmultiplayer-ai-and-multi-agent-ai\u002F","Multiplayer AI versus multi-agent AI"," is the distinction between that room and a cast of models.",[821,1699],{},[206,1701,826],{"id":825},[528,1703,1704,1709,1714,1719],{},[317,1705,1706],{},[461,1707,1708],{"href":1344},"McKinsey, The State of AI: Global Survey 2025",[317,1710,1711],{},[461,1712,1713],{"href":463},"Microsoft and LinkedIn, 2024 Work Trend Index",[317,1715,1716],{},[461,1717,1718],{"href":1382},"Yang et al., Nature Human Behaviour, 2022",[317,1720,1721],{},[461,1722,1723],{"href":1471},"Melvin Conway, How Do Committees Invent?, 1968",{"title":169,"searchDepth":170,"depth":170,"links":1725},[1726,1727,1728,1729,1730,1731,1732],{"id":1329,"depth":170,"text":1330},{"id":1355,"depth":170,"text":1356},{"id":1441,"depth":170,"text":1442},{"id":1558,"depth":170,"text":1559},{"id":1577,"depth":170,"text":1578},{"id":1667,"depth":170,"text":1668},{"id":825,"depth":170,"text":826},"A personal assistant is not team software. Solo AI is one person and one chat. Multiplayer AI is a job more than one person can open — same brief, same files, same refusal.",{"eyebrow":918,"title":1735},"Solo chat versus a shared room",[1737,1740,1743,1746],{"question":1738,"answer":1739},"What is the core difference between Solo AI and Multiplayer AI?","Solo AI connects one human user to an isolated AI chat session, requiring manual copy-pasting of outputs. Multiplayer AI enables multiple human users and autonomous AI agents to work together simultaneously inside a shared visual or textual environment with real-time state synchronization.",{"question":1741,"answer":1742},"How does Multiplayer AI handle editing conflicts between humans and AI?","Multiplayer AI uses mathematical synchronization protocols known as Conflict-Free Replicated Data Types. These systems treat human typing and streaming AI token updates as mergeable operations, preventing data loss, text overwrite, or cursor jumping during simultaneous editing.",{"question":1744,"answer":1745},"Is Multiplayer AI secure for confidential enterprise data?","Yes. Enterprise multiplayer AI puts role-based access and guardrails in the agent architecture. An agent inherits the privileges of the person who invoked it, or the permissions of the workspace it is operating in. It does not receive a broader grant than that person or that room.",{"question":1747,"answer":1748},"Can existing software tools be upgraded to Multiplayer AI?","Yes. Traditional platforms can transition by integrating real-time document sync engines, deploying shared semantic memory layers, and replacing standard forms with multi-user canvases supporting background AI agent workers.","\u002Fblog\u002Fsolo-ai-vs-multiplayer-ai",{"title":1248,"description":1733},"blog\u002Fsolo-ai-vs-multiplayer-ai",[1239,1242,1244,1243],"wj3A1cwOHBroMNu9t6gT3V1iMQfDIdeT0JNRjsZlS3I",{"enabled":163,"message":1755,"linkLabel":79,"linkHref":80,"id":1756,"title":1757,"archived":163,"authors":164,"badge":164,"body":1758,"date":164,"definedTerm":164,"department":164,"description":169,"extension":172,"eyebrow":164,"faqHeader":164,"faqs":164,"footerBand":164,"headline":164,"image":164,"industry":164,"jobType":164,"listed":131,"location":164,"navigation":131,"openRoles":164,"pageLayout":164,"path":1762,"relatedHeading":164,"seo":1763,"series":164,"sitemap":163,"status":164,"stem":1764,"subhead":164,"tags":164,"video":164,"whyJoin":164,"workplaceType":164,"__hash__":1765},"We're hiring! Join the team building the Sentient Enterprise.","content\u002Fshared\u002Fhiring.md","Hiring banner",{"type":166,"value":1759,"toc":1760},[],{"title":169,"searchDepth":170,"depth":170,"links":1761},[],"\u002Fshared\u002Fhiring",{"title":1757,"description":169},"shared\u002Fhiring","1zs3boivKda1e-b-hAyuNcmZSKjZUAXmecnwHVgcHzk",{"fold":1767,"id":1771,"title":1772,"archived":163,"authors":164,"badge":164,"body":1773,"date":164,"definedTerm":164,"department":164,"description":169,"extension":172,"eyebrow":164,"faqHeader":164,"faqs":164,"footerBand":1777,"headline":164,"image":164,"industry":164,"jobType":164,"listed":131,"location":164,"navigation":131,"openRoles":164,"pageLayout":164,"path":1781,"relatedHeading":164,"seo":1782,"series":164,"sitemap":163,"status":164,"stem":1783,"subhead":164,"tags":164,"video":164,"whyJoin":164,"workplaceType":164,"__hash__":1784},{"headline":1768,"description":1769,"primaryLabel":8,"primaryTo":1770,"secondaryLabel":956,"secondaryTo":12},"Run frontier AI your business actually owns.","Governed workstreams, 3,000+ integrations, and a proprietary knowledge graph. Start on Free.","\u002Fsignup?plan=free","content\u002Fshared\u002Fcta.md","Site CTAs",{"type":166,"value":1774,"toc":1775},[],{"title":169,"searchDepth":170,"depth":170,"links":1776},[],{"headline":1778,"description":1779,"primaryLabel":8,"primaryTo":1770,"secondaryLabel":1780,"secondaryTo":85},"See what governed AI looks like on your stack.","Connect your tools, run a workstream, and keep every decision on your ledger. Start on Free.","Talk to our team","\u002Fshared\u002Fcta",{"title":1772,"description":169},"shared\u002Fcta","eYqahyaPnbp8GKrWpoORbZdtmkWmgHr5F61ZHOnb8sY",1791647069896]