Loading
Loading
Bob walks the customer through the claim, reads the evidence, checks the policy, and prepares a routine payout for a licensed adviser to approve. It fast-tracks what is clearly covered and escalates anything that needs a human. Bob never settles or pays on its own.
Claims are the moment of truth for trust. Here is the whole path, and where a licensed human always steps in before anyone is paid.
Bob runs a warm, plain-language conversation the moment something goes wrong, on your site, chat, or email, in your brand. It discloses it is AI and tells the customer a person approves before any payment.
In your brand, day or night
Bob reads the photos and documents the customer sends, matches them against the cover on file, and works out what looks payable. Never a figure without its working shown.
Checked against the policy
For a claim that clearly matches the cover and sits under your firm's limits, Bob prepares an itemised payout with a confidence level and a fast-track flag. Anything unusual is escalated, fully briefed.
Fast-track the routine
The assessment stays prepared, awaiting approval. One of your team reviews it, adjusts if needed, and approves the payment. Bob never settles or pays on its own, and the whole thread is on the record.
A human approves every payout
A general chatbot cannot read a receipt or your policy, so it cannot itemise a claim. Bob does the opposite: the prepared payout is built from lines you can see, each matched to a benefit, with a confidence level and the evidence behind it.
An example of a claim on your own site, in your brand. Bob discloses it is AI, prepares the routine payout, and hands the payment decision to a licensed person.
The customer feels looked after at a stressful moment. Your team gets a claim that is already assessed and easy to decide.
A general chatbot cannot read the evidence or the policy, so it cannot prepare a real assessment. Manual claims are trustworthy but slow. Bob keeps the speed on the routine and the accountability on every payout.
| A general chatbot | Bob | Manual claims | |
|---|---|---|---|
| Reads photos and itemises the claim | No | Yes | By hand, slower |
| Checks the policy and cites the benefit | No | Yes | Manual |
| Fast-tracks routine, escalates the rest | No | Yes | No |
| Adverse outcomes get a licensed sign-off | No | Yes | Yes |
| Every step logged for audit | No | Yes | Manual |
| Settles or pays on its own | No | No, prepares for a human | The assessor decides |
Bob starts as a copilot. It prepares and explains; your team reviews and approves. You turn up autonomy per action as trust builds, within limits you set, and adverse outcomes are never auto-dialable.
Every assessment is clearly labelled prepared, awaiting approval. Bob reads the evidence, checks the policy, and fast-tracks the routine, but it never settles or pays on its own. A licensed person reviews and approves before any money moves. Any decline, partial payout, or exclusion is an adverse decision that always needs a licensed sign-off, never an automated one, and a disputed claim is routed to an independent complaints desk on the Fair Insurance Code clock. Every step is logged with reason codes and cited sources.
No. Bob prepares a routine claim: it reads the evidence, checks the policy, and itemises what looks payable, all clearly labelled awaiting approval. A licensed person reviews and approves before any money moves. Bob never settles or pays.