Most AI tools bolt a chatbot on top of an app. Nahla is the other way around: the data model, the governance rules, and the permissions came first. The AI sits on top of all three, and cannot reach outside them.
Natural-language chat and agents. Every answer is grounded in the governance rules and data below.
| A generic AI bolt-on | Nahla, by design | |
|---|---|---|
| Your schedule | Paste full schedule = data leak | Stays in your workspace |
| .XER / .MPP | Can't read XER / MPP natively | Reads XER / MPP natively |
| Dates & logic | Hallucinates dates & logic | AI on governance rules, not guesses |
| CPM / DCMA | No CPM / DCMA | DCMA-14 built in |
| Approval | No approval step · no record of who approved what | Every change approved and recorded |
The boundary is in code, not in a prompt. The AI has a fixed list of tools. Every agent run, proposal, approval and outbound email is recorded.
Drafts a change, you approve
Computes, never changes
Never, ever
Every agent, including the scheduled ones that run in the background, only proposes. Nothing reaches your schedule, and no email reaches a subcontractor, until you approve it.
Chat and background agents draft typed proposals. The only way a change lands in your schedule is your explicit approval, through the same single write path.
Progress Collection and Weekly Report draft into your inbox. Emails go out only when you approve them, and only to the recipients you choose, never automatically.
Agents use the same tool allowlist as the AI rail. Governance rules calculate every number; baselines stay read-only; every agent run and approval is recorded.
Every change the AI proposes waits for your explicit approval on a single write path. Approve it and it applies; change your mind and step back with undo in Build, up to 50 steps.
nahlaaiset_durationIM.2.2 Steel Erection · add_linkIM.2.3 → IM.2.4 · level_resources · appliedFoundation Crew · shift May 15‑28 → Jun 3‑7Undo in Build, up to 50 steps · Ctrl+Z reverts the last applied change.
Hosted on AWS in Sydney and separated by account.
Hosted in AWS Sydney on fully managed infrastructure that supports GDPR compliance.
Saved in your project so you can pick them up again, and deleted with it.
Strict boundaries between accounts. Every request authenticated and access-checked.
P6- and MSP-ready, CSV and PDF export. One click to delete the account.
No selling. No reselling. Delete on demand, including your full history.
Tasks, links, calendars, resources and baselines all have strict schemas. The AI cannot invent fields. Validation runs before any change is committed, and baselines are read-only to the AI.
CPM and DCMA-14 are exact algorithms, not language-model guesses. The AI applies the rules for results, then quotes the numbers verbatim. Same input, same answer, every time.
A generic chatbot needs your whole schedule pasted into a public chat. With Nahla it stays in your isolated workspace, the AI sees only what a question needs, and nothing is used to train AI models.
P6- and MSP-ready, CSV and PDF export at any time. One click wipes your projects, history and account.
We'll send our data-security brief, and a DPA on request.