An assistant you command in plain Arabic or English — and an agent that watches, judges and acts on its own. Not a chatbot on a dashboard: intelligence wired through every module.
A workspace-embedded AI that interprets what you mean and does it — search, replay, reports, map control, exports.
Continuous fleet review, 24/7. It opens cases, scores severity, proposes actions — and takes them, under your rules.
Van 1187 starts moving — no assignment, outside working hours. The agent opens an unauthorized-use case.
Cross-check: no booking, no assigned driver shift. Severity: Critical.
The manager's phone rings with the alert: live location and a replay link, while the vehicle is still inside the city.
Vehicle stopped and recovered at the gate. Case annotated, evidence preserved.
Nine minutes from ignition to intercepted. That's the agent.
Severity badge, location snippet, the evidence curve, and one approve button. The agent doesn't send you raw data — it sends you a decision that takes five seconds to make.
Approve it, reject it, or let your rules answer for you.
Every issue becomes a case driven from open to resolved — you manage exceptions, not screens.
Continuous fleet review scans every vehicle and event, around the clock.
Anything off-pattern opens a case automatically, the moment it happens.
AI investigates context and scores severity from Low to Critical.
The right action is proposed — approved by you, or auto-approved by your rules.
Driver notified or manager escalated, then tracked to closure.
Dial the agent's independence up at your own pace. It earns trust before it earns authority.
Start observe-only: it opens and assesses cases without acting — so you can verify its judgment first.
Maker-checker: the agent proposes, a manager approves or rejects before anything is sent.
Auto-act only above the level you choose; everything below is logged and reported.
Open, assess, propose, approve, send, escalate, resolve — every step, timestamped and reviewable.
Inside the dashboard, not in a separate tab. It opens replays on the real map, drops report files into the chat, and hands you next-action chips after every answer.
Continuous speech recognition with spoken replies — Arabic and English, auto-detected. The mic listens, the AI resolves the command, the map and reports respond, and it answers out loud.
Built for dispatch rooms where hands are busy and minutes matter.
No. Your data stays in your workspace and is used to serve your fleet — the assistant and agent read it to answer you and manage cases, nothing more.
It's a first-class citizen: voice and text, dialect-tolerant, auto-detected, with answers in the language you used. Even ordinals like "السيارة رقم ٣" resolve to the right vehicle.
That's what the guardrails are for: start in shadow mode and check its judgment, keep maker-checker approval on actions, and review the audit trail. You expand its authority only as it proves itself.
Every tier — from Personal to Enterprise — has an AI flavor that includes the assistant, voice and agent, with monthly usage quotas sized per tier. See pricing.
Yes — the 14-day trial includes the AI. Connect one vehicle and talk to it the same day.
In the demo, you'll talk to a live fleet — in Arabic or English — and watch the agent open a case in front of you.