Skip to content

How DutyGraph uses AI ​

DutyGraph uses AI to draft, read and suggest. It never uses AI to decide. Every AI output is a proposal that waits for a person, and everything that grants, refuses or records authority is code.

Where AI is called ​

The Key column says which connection pays for and receives the call:

  • Company: the company's approved Claude connection first, then its approved OpenAI connection, else no call.
  • Company (OpenAI): the company's approved OpenAI connection only.
  • Firm (OpenAI): the OpenAI key in Settings › Research & AI keys. A person with shared access to one company uses that company's approved OpenAI connection instead.
  • Platform: a key DutyGraph holds on the server.
WhereWhat the AI doesKeyWhat a person does
Company researchReads public pages from the research provider and extracts field values. Each quote is checked against the page.Company (OpenAI)Findings stay Inferred until a client person confirms them
Research evidence checks (Jev)Answers typed questions about evidence and outreach text, as probabilities. Code sets every threshold and can only make a result stricter.OpenAI, or TypeSafe on a platform key if the firm opts inOptional. Without it, rules decide.
Here's what we heard on the sponsor packet and team asksTidies each answer for meaning and adds at most one suggested improvementCompanyThe person confirms or corrects it before sending
Companion phone appTranscribes a voice answer and drafts the understanding of itPlatform (xAI)The person checks it before sending
Voice transcription in the browserTurns a recording into editable textFirm (OpenAI)The person edits the text before sending
Voice interviewsRuns a live spoken interview and drafts a summaryCompany (OpenAI)The advisor reviews the transcript. The summary is a labelled draft.
Executive kickoffPrepare meeting guide drafts an 8-topic guide. Prepare team review reads the pasted notes or transcript and drafts people, roles, managers and duties.Firm (OpenAI)The advisor reviews every proposed person and duty
Kickoff correctionsGenerate AI correction proposals turns the sponsor's comments into proposed changesFirm (OpenAI)The advisor accepts or rejects each one
Team asksCreate question sets writes a tailored set of questions for each person from their recorded dutiesFirm (OpenAI)The advisor edits and sends them
Task CardsDraft task cards turns accepted answers into Task CardsFirm (OpenAI)Each card is Proposed and Inferred until accepted
Repo fill loopReads answers and meeting notes into field readings and proposed duties, Task Cards, handoffs and workflowsCompanyEach reading waits in Repo › To review
Continuing checksReads a reply to a follow-up question and proposes work updatesFirm (OpenAI)Proposals never overwrite reviewed or confirmed work
Work MapConnect existing work with AI proposes stage links for older workFirm (OpenAI)Links stay dashed until a person accepts them
Team analysisAnalyze team work suggests cross-team hypotheses with exact citationsFirm (OpenAI)The advisor decides on each one
DutyGraph AgentAnswers, reviews, proposes edits and drafts asks. On the Strategy pages it runs in Ask mode.CompanyThe advisor accepts or rejects each edit. Asks queue unsent.
Outcomes and StrategyFind possible constraints and the RACI, Theory of Constraints and SWOT / TOWS framework runsCompanyThe advisor accepts or rejects each result. Code sets the evidence words.
Route RefreshStarts a research run for the AI patterns that fit the companyAs company researchEach option waits in review
Agents › Business reviewAn optional AI draft of a manifest for an older requestFirm (OpenAI)Requested scopes remain proposals

Where code decides, with no AI ​

AreaWhat code does
Hours a weekReads the tapped range, such as "5–10 hours"
AI-fit passProposes an AI mode, owner and performer for each Task Card, with a plain reason
Proposed agentsGroups Task Cards into agents for one named person, inside that person's access
RouteEvery verdict, fit, value and option order
Constraint signalsRules that point to Repo files
Company Repo, Work Map and reportsCompiled from records. The same records give the same output.
Kickoff agenda and Ask the person responsibleBuilt from fixed templates and open fields
Agent builderMarkers, the AI mode cap, tool policies and the five export formats
Agent plan pageBuilt from the saved manifest
The key joinThe governance evaluator gives each requested scope one state with a rule citation
Approval chainRecords each person's decision. No call from an AI chat can decide.
LedgerHash-links each decision
GatewayAllows or blocks each agent call routed through it

The rules ​

  1. AI only proposes. AI can save a draft or a proposal. It never becomes an accepted fact, a confirmed record or a permission until a person accepts it.
  2. An approved connection, never a borrowed one. A company's AI calls go through a key the firm saved or the company approved. On the hosted service, the only AI keys DutyGraph holds itself are the platform keys marked above. Saved keys are encrypted and never sent to the browser.
  3. Sample companies. Features on the Company key, company research and Route Refresh never call a provider for a sample company. They show Sample output · no AI call or simulated research.
  4. Accepted values in, raw words out. DutyGraph Agent and Strategy read compiled Repo files: accepted values only, with email addresses masked. A person's raw words never reach a DutyGraph Agent edit or a drafted ask. Holdings never reach any AI.
  5. No stored prompts. Usage rows keep counts, sizes, the model and the cost, not the message. Text calls to OpenAI and xAI set store: false.
  6. AI input retention. A job that runs at most once an hour removes quotes from stored Strategy runs when the quoted file changes or leaves the Repo. It also removes evidence text from older framework runs and team analyses. See Security and data handling.
  7. Started on purpose. Most AI actions start when a person selects them, after a consent note where the page shows one. The fill loop and continuing checks run as background jobs when an answer arrives. A replay of the same command never makes a second call.

Four meanings that stay apart ​

AI proposals, participant confirmation, advisor approval and external authority mean different things:

KindWhoWhat it means
AI proposalA modelA draft. It changes nothing.
Participant confirmationThe person who answered"This is what I meant." It does not make it company practice.
Advisor approvalThe advisorThe advisor accepted it into the company's record, or approved a report for an audience.
External authorityPeople in the company who already approve accessA decision in the approval chain about a governed agent

One never stands in for another. A confirmed answer is not an approval. An AI reading stays a proposal until the right person validates it. An approval in DutyGraph does not change a business system: the company's IAM provisions access. See Approval, provisioning and revalidation. See Evidence and provenance.

Limits ​

  • Several older features marked Firm (OpenAI) do not check for a sample company on the server. On a sample company they can call OpenAI if the firm saved an OpenAI key.
  • No live AI provider is called in the automated tests. They use fake providers.

INFO

Status. Built: every AI call in the first table, each through the key shown, and every code-only step in the second table. Simulated: research for sample companies uses labelled fixtures. Planned: an optional, labelled AI summary paragraph on reports.

Source: gitbook/product/architecture/how-dutygraph-uses-ai.md in the DutyGraph repository.

DutyGraph. Pages label each feature Built, In review or Planned.