Every Suspicious Motor Claim Can Now Go to a State SIT: What a Commercial Fleet Must Keep on File

In Oriental Insurance Co. Ltd. v. Tuni Pati, 2026 SCC OnLine SC 1710, decided on 17 August 2026, a bench of Justices Ahsanuddin Amanullah and Prasanna B. Varale directed all states to establish dedicated Special Investigation Teams for investigating suspected fraudulent insurance claims. The direction is addressed to the states, so the machinery is police machinery, not an insurer's internal investigation cell.

The second direction is the one that changes daily practice. Insurers were required to forward all potentially fraudulent claims to the relevant state SIT, without selective filtering, and the Court placed accountability for selective reporting on top management. An insurer that decides a suspect claim is too small, too old or too commercially awkward to refer is no longer making a routine commercial judgement.

The Court also impleaded IRDAI, the Ministry of Finance, the Ministry of Road Transport and Highways and the General Insurance Council, and proposed a common portal integrating VAHAN, SARATHI and e-DAR for cross-verification of repeat claimants. The machinery is being settled with the regulator, the road transport ministry and the industry body at the table, which is why the detail below is about what a fleet controls rather than what is still being drafted.

For an operator running trucks, buses, tankers or last-mile fleets, the practical effect is that an ordinary tribunal claim can now be routed into a police investigation on the insurer's say-so, and the insurer has been told it may not hold that referral back.

Insurers already maintained fraud units and already referred a small number of egregious cases to the police. What they did with the rest was a commercial calculation: cost of investigation against quantum at stake, relationship with a fleet account, the prospect of a negotiated settlement before the tribunal. That calculation is what the Court has removed.

Once referral becomes the default for anything the insurer flags as potentially fraudulent, the population of referred claims stops being the small tail of obvious rackets. It starts to include the ordinary claim with a defective file: the FIR registered four days late because the driver called the transport manager before he called the police, the licence whose transport endorsement had lapsed on the date of the accident, the trip sheet that shows a route the telematics unit does not corroborate.

The defensive posture is therefore not about disproving fraud after the fact. It is about making the file so internally consistent at intimation that nothing on it reads as a flag in the first place. Fleets that already run disciplined documentation for cargo and transit claims will recognise the discipline; motor third-party files have historically been held to a looser standard, and that gap is what closes now.

The portal the Court proposed would join three national systems that already exist separately.

• VAHAN holds vehicle registration data: owner, class of vehicle, fitness, permit and registration status.
• SARATHI holds driving licence data: licence number, validity, class of vehicle authorised, and the transport endorsement history.
• e-DAR is the Ministry of Road Transport and Highways electronic repository of accident reports, into which the police-generated accident report for a tribunal claim is filed.

Individually, each is queried today on a claim-by-claim basis. Joined and queried together, they answer a different question: which registration numbers, which licence numbers and which claimants recur across accident reports. That is the repeat claimant cross-verification the Court described, and it is a pattern query rather than a claim query.

A fleet sits on the wrong side of that query by construction. A single owner runs dozens or hundreds of registrations, employs drivers whose licences appear on multiple vehicles, and generates more accident reports in a year than any individual owner will in a lifetime. Legitimate frequency looks like frequency. What separates a clean fleet from a flagged one is whether each individual file survives inspection when the pattern query pulls it up, so the useful investment is in per-claim file quality rather than in trying to suppress the frequency signal.

Licence defects are the most common and most easily proved insurer defence in motor claims, and they are exactly what a SARATHI-linked query surfaces first. Three failures recur in fleet files.

1. Class mismatch. The driver holds a licence for a light motor vehicle and is driving a goods carriage above that threshold, or holds a goods endorsement and is driving a passenger vehicle.
2. Lapsed transport endorsement. The base licence is live but the transport authorisation, which is renewed on a shorter cycle, expired before the date of the accident.
3. Renewal gap. The licence was renewed after expiry, and the accident falls inside the gap.

The operational fix is a scheduled SARATHI re-verification for every driver on the roster, with the query result and its date stored against the driver record, plus a hard block that prevents dispatch on an expired or mismatched authorisation. When the claim file can show a verification run dated shortly before the accident, the licence line of enquiry closes quickly. When it cannot, the insurer has both a coverage defence and a referral trigger, and a third-party liability award that the insurer pays under a pay-and-recover direction becomes a recovery action against the fleet.

Delay in registering the first information report is the single most cited indicator in fraud investigations, because a delayed FIR is what a staged or transferred accident usually produces. The delay itself is rarely fatal to a genuine claim, and courts have long accepted that an injured driver, a rural location or an uncooperative station explains it. The problem arises when the delay is unexplained on the file.

Under the detailed accident report scheme introduced by the Central Motor Vehicles (Fourth Amendment) Rules, 2022, the police generate a sequence of reports for the Claims Tribunal: a first accident report, an interim accident report, and the Detailed Accident Report that carries the investigating officer's findings on the vehicles, drivers, licences, insurance particulars and the manner of the accident. The DAR is the document the tribunal works from, and it is the document that flows into e-DAR.

That gives a fleet one specific obligation it usually neglects. The DAR is prepared by the police from what they are given, and once filed it becomes the baseline narrative. If the vehicle, driver, load and sequence recorded in the DAR differ from what the fleet later states to the surveyor or the tribunal, the inconsistency is on the record before the fleet is even aware of it. A transport manager who obtains a copy of the DAR when it is filed, reads it against the internal incident report, and raises any discrepancy in writing at that point is doing the highest-value hour of work in the entire claim.

The timing sequence a clean file shows is simple: FIR registered the same day where the driver is fit to report, an internal incident report raised on the same date, intimation to the insurer within the policy notification period, and a DAR whose facts match all three. Where any leg is late, the file should carry a contemporaneous explanation rather than a reconstruction offered eighteen months later.

The vehicle location tracking device fitted under the AIS-140 standard, mandatory for public service vehicles and national permit goods carriers, produces the only independent, timestamped record of where a vehicle was and how fast it was moving. In a fraud enquiry it is the evidence that settles the point in whichever direction the facts run.

A genuine claim is corroborated in minutes: the position trace puts the vehicle at the accident location at the time in the FIR, the speed profile shows the deceleration, and the trip matches the consignment and the trip sheet. A staged claim collapses just as fast, because the trace cannot be back-dated once it has been transmitted to the backend.

The failure mode is not the device. It is retention. Many fleets subscribe to a tracking platform that retains raw position data for ninety days or less, while a tribunal claim runs for years and an SIT referral can arrive long after the underlying trip has rolled out of the platform's window. The claim file should carry an exported, timestamped extract of the relevant trip pulled at intimation, not a promise that the data exists somewhere upstream. The same discipline applies to dashcam footage and panic-button events, and the argument runs in detail in our note on VLTD, panic buttons and the telematics log in transit claims.

Assemble this set at intimation, not when the insurer's investigator calls. Everything on it is available on day one and gets harder to obtain with every week that passes.

1. Vehicle: registration certificate, fitness certificate, permit and the VAHAN status printout as at the date of the accident.
2. Driver: licence with class and transport endorsement, the dated SARATHI verification result, employment record and duty roster showing the driver was on assigned duty.
3. Incident: FIR copy with the registration date and time, the internal incident report, driver statement recorded within twenty-four hours, and photographs of the vehicle and site.
4. Movement: VLTD extract for the trip, trip sheet, consignment note or e-way bill, and any dashcam or panic-button record.
5. Policy: the certificate of insurance in force, the intimation acknowledgement with its timestamp, and the surveyor appointment reference.
6. Follow-through: the DAR copy when filed, with a written note reconciling it against the internal report.

Ownership matters as much as content. In most fleets the branch manager assembles motor files, the safety head owns telematics and HR owns licences, so no single person can produce the set. Name one custodian per claim, give that person read access to all three systems, and set the deadline at forty-eight hours from intimation. Fleets that treat this as an extension of their existing commercial motor and fleet insurance administration find most of the inputs already exist and are simply not collated. The same file also improves what the fleet can negotiate at renewal on its motor insurance programme, because a claims history backed by clean documentation defends its own loss ratio.

The 17 August order set the direction and left the machinery to be worked out with the parties the Court impleaded. Four things stay open, and each of them changes how much weight a fleet's own file has to carry.

• Who builds and governs the portal. VAHAN, SARATHI and e-DAR all sit with MoRTH, while claims data sits with insurers under IRDAI. Access rules, retention periods and whether a fleet or a claimant can see and correct its own record are unresolved.
• What counts as potentially fraudulent. Absent a definition, insurers referring everything to avoid the accountability exposure would swamp state SITs and slow genuine claims. A threshold or an agreed set of indicators would move referral volumes sharply in either direction.
• How a fraud-based repudiation is evidenced. An insurer that refers a claim to an SIT and then repudiates on fraud grounds is relying on investigation material the policyholder has not seen. Whether the fleet gets sight of it, and at what stage, is not answered by the order.
• What the General Insurance Council standardises. Industry-level referral protocols and a common evidence checklist would let fleets prepare to one known standard rather than to each insurer's practice.

None of this changes what a fleet should do in the meantime. Referral is already the stated default, the three databases already exist and are already queried claim by claim, and the documents that answer a fraud enquiry are documents that have to be captured on the day of the accident. The parallel tightening on the pricing and settlement side is covered in our note on IRDAI motor third-party reform and commercial fleets.