Insurance · Claims Fraud

Fraud caught before payout. Not during quarterly review.

Document forensics, image tampering detection, claimant + repairer network graphs, and similarity search over 4M+ historical claims — wired into your claims system, explainable to your SIU, defensible to your regulator.

claims-fraud · sentinelLIVE
CLAIM · CLM-40821
Auto claim · staged collision network
PAYOUT
$218,400
7 claimants
FRAUD SIGNALS
Claimant networkpending…
Cluster centralitypending…
Similarity · 2019 ringpending…
Invoice patternpending…
SIU precedent hitpending…
FRAUD MODEL SCORE
0.00 · clean0.50 · review0.70+ · fraud
REASONING
Network graph matches 2019 staged-ring pattern
Repairer collusion across 3 claimants
Invoice parts pricing 38% above benchmark
SCREENING…
SIU referral · case opened

What we build

A fraud stack that sees what single-claim scoring misses.

Every capability is a production layer — documents, images, graph, precedent — working together on the same claim in the same second, so nothing escapes because two signals were siloed.

Document-pattern anomaly detection

Invoice tampering, date inconsistencies, provider-signature forgery, and copy-paste narrative patterns flagged across every FNOL, police report, and invoice before the claim advances.

Image-tampering forensics

EXIF metadata cross-checked against device + time + location. Pixel-level computer vision catches splicing, re-saves, and synthetic damage in every scene photo.

Claimant + repairer network graphs

Entities resolved across claims history into a live graph. Centrality, clustering, and motif matching surface collusion rings invisible to per-claim scoring.

Similarity search over 4M+ claims

Vector similarity finds every historical claim that looks like today's — same narrative, same repairer, same staged pattern — so precedent becomes evidence instantly.

Soft-fraud vs hard-fraud scoring

Exaggeration, padding, and opportunism scored separately from organised ring fraud — so treatment (adjust) and escalation (block + SAR) match the risk class.

SIU case management + SAR filing

Every flagged claim arrives in the special-investigation queue with evidence pack, draft narrative, and SAR-ready exports — so investigators spend time on decisions, not assembly.

Where the stack catches fraud

Every line of business. Every fraud class.

Same document, image, and graph layers — tuned per line of business. Shared entity resolution, per-LoB models, cross-line ring detection. The patterns below all run on the same stack; only the models, thresholds, and integrations change.

01

Auto · staged accidents

Motor fraud rings, collusive repairers, phantom passengers, and cash-for-crash schemes — caught by graph centrality and similarity to known historic rings.

02

Health · invoice inflation

Overlapping treatment dates, phantom visits, upcoded procedures, and provider-watchlist hits — surfaced by doc-pattern models and provider anomaly scoring.

03

Property · staged loss

Claim amounts inconsistent with weather events, pre-loss imagery mismatches, and suspicious contractor networks flagged before the inspector is dispatched.

04

Specialty · commercial lines

Cargo, marine, and commercial-property claims scored against historical ring precedent and counterparty network — higher-value, higher-asymmetry decisions.

05

Organised ring detection

Cross-policy, cross-LoB entity resolution — the same ring operating across motor, health, and property surfaces as a single network, not three isolated cases.

06

Soft-fraud · exaggeration

Opportunistic padding of legitimate claims — detected, sized, and adjusted without blowing up the customer relationship or dragging the SIU in unnecessarily.

A walk-through

From claim landing to SIU case file — five clear layers.

Follow a real motor claim through ingest, forensics, network graph, flagging, and SIU hand-off. Every step is visible to the adjuster, the investigator, and the regulator.

ANCHOR CARRIER · SIU WORKFLOW
Sentinel Assurance Pte Ltd· SG/MY motor + health · 22k claims/mo · 18 investigators
STEP 01 · 05
STEP 01 · INGEST
Every artefact, in one stream
FNOL, invoices, images, police reports, and claimant history pulled into a single case record the moment the claim lands.
New claim landed
Streamed into the unified case record
Case IDCLM-40821
Line of businessMotor · comprehensive
Intake channelAgent portal · 04:22 SGT
Sources merged9 · zero gaps
ARTEFACTS INGESTED
FNOL report
parsed · 48 fields
Invoices × 4
OCR · confidence 0.94
Scene images × 7
EXIF captured
Claimant history
2 prior flags
Case file assembled in 2.1s · ready for analysis

Model families we deploy

No single model catches every fraud. So we ensemble.

Each model family targets a distinct fraud surface — documents, images, networks, precedent. Blending their outputs is what separates catching a staged ring from catching a single inflated invoice.

NLP + STRUCTURAL
Doc-pattern anomaly detector

Gradient-boosted + transformer ensemble over invoice structure, narrative style, date logic, and provider metadata. Tuned to the tampering signatures the team has seen before.

EXIF + PIXEL FORENSICS
Image-tampering CV

Dual-pipeline — EXIF + device-attribution check alongside error-level-analysis and noise-residual CV. Catches re-saves, splicing, and AI-generated imagery.

CENTRALITY + MOTIF
Network-graph GNN

Graph-neural-network over the claimant-repairer-provider network — surfaces ring-like motifs, shared-address clusters, and centrality spikes before the 4th claim lands.

VECTOR RETRIEVAL
Similarity search · FAISS

Dense embeddings over narrative + metadata + graph neighbourhood. FAISS retrieves the closest historical claims in milliseconds — precedent becomes part of the decision.

Data sources wired into every model

Every signal that moves the verdict — integrated.

Pulled in parallel, normalised into a single case schema, versioned alongside the models that consume them.

Source
What it unlocks
Coverage
Claim documents
Every document parsed at intake — OCR + structural + narrative analysis feeding the document-pattern model and evidence vault.
FNOLInvoicesPolice reportsMedical recordsRepair estimates
Image metadata + pixels
Scene imagery analysed beyond what the human eye sees — tamper detection, AI-generation detection, and device-attribution on every photo.
EXIFDevice fingerprintELANoise residualGeotag
Claimant history
Full claimant profile across policies and incidents — prior fraud flags, disputed claims, and pattern signals contributing to today's decision.
Policy adminPrior claimsSIU flagsPayout historyComplaints
Repairer network database
Every repairer, clinic, and contractor is a node — pricing norms, volume fingerprints, and regulatory sanctions feed the graph layer directly.
Approved vendorsWatchlistsPricing benchmarksVolume patternsInvoice archives
Historical claim archive
4 million historical claims indexed for similarity retrieval — precedent isn't a lookup chore, it's a live feature of every new decision.
4M+ claimsVector embeddingsNarrative corpusSAR archiveAdjuster notes
External fraud watchlists
Integrated with industry-shared and regulator-sourced fraud lists — adds precedent the carrier hasn't seen internally, without replacing proprietary signals.
Coalition AIFFRISSShift NetworkRegulator bulletinsIndustry shared data

Explainability, not just flags

A flag alone doesn't pass SIU review. An evidence pack does.

Every block, investigate, or clean-through is accompanied by top-feature reasoning, document + image + graph evidence artefacts, model-version provenance, and a draft SAR narrative — generated at decision time, indexed for audit.

  • Top-feature contributions per flag (document + image + graph)
  • Full evidence pack with artefacts and historical matches
  • Draft SAR narrative and regulator-template export
  • Aligned to MAS, NAIC, FATF, Coalition AIF
SIU PACKET · CLM-40821
fraud.explain v4.1
VerdictBLOCK · SIU referral
Top signal 1graph_centrality · 0.36
Top signal 2doc_pattern · 0.29
Top signal 3ring_similarity · 0.22
Model pathdoc v4 + img v3 + gnn v2
SAR templateFATF-SG-2026
Audit SHAa91f…e230

Governance & audit

Defensible to SIU, compliance, and the regulator.

Every block, investigate, or SAR filing arrives with the provenance, reasoning, and due-process trail your internal audit and external regulator both expect — not bolted on at the end of the quarter.

Every point ships with the detection stack. Audit-ready from day one.

Model explainability per flag

Every fraud flag produces top-feature reasoning — document signal, image signal, graph signal, historical match — so SIU and compliance see why before they act.

False-positive rate disclosure

FPR tracked per model, per LoB, per cohort — reported back to the business monthly. Threshold changes happen only when the trade-off is documented.

SAR filing audit chain

Every Suspicious Activity Report traces back to the claim, the signals, the investigator, and the model version — defensible under regulator inspection.

Claimant due-process notifications

Statutory due-process letters, appeal paths, and evidence disclosures generated alongside the block — so customers keep their rights even when the claim doesn't advance.

Anti-bias review on demographics

Outcome-equity analysis across age, geography, and postcode proxies on every model deploy — so fraud scoring doesn't quietly discriminate.

Investigator workflow audit

Every SIU action — override, escalate, close, archive — logged immutably. Adjuster and investigator activity visible for internal audit and external inspection.

Frameworks we align to

IFRS 17NAIC fraud rulesMAS Insurance ActFATFGDPRISO 27001ACORDCoalition Against Insurance Fraud

Why Axccelerate for claims fraud

Not a rules engine.
A detection stack.

A rules engine fires on what you already knew. Our stack surfaces patterns hidden across documents, images, networks, and precedent — the fraud your SIU team hasn't seen yet.

Feature
Axccelerate
Legacy fraud vendor
In-house rules
Document + image + graph in one stack
Varies
4M+ historical claim similarity search
Varies
Soft vs hard fraud separation
Varies
SIU workflow + SAR generation
Varies
Varies
Image EXIF + pixel-level tamper forensics
Varies
Network-graph GNN (not just rules)
Claimant due-process notifications
Varies
Anti-bias outcome-equity review
Varies
Shared-industry + regulator watchlist feeds
Varies
No vendor lock-in · proprietary model output

Pricing

Priced to the LoB + volume, not per flag.

Fraud deployments are custom — we scope against your lines of business, claim volume, and SIU tooling before quoting.

Launch
Enquirefor pricing
Single LoB fraud model

One line of business — motor, health, or property — with document + image forensics and a starting graph. Integrated to your claims system and SIU tooling.

1 LoB fraud model (ensemble)
Doc + image + graph layer
SIU case routing + evidence pack
Monthly FPR + drift reporting
InsightAX reporting access
Enquire for pricing
Most popular
Scale
Enquirefor pricing
Multi-LoB carrier

Fraud scoring across motor, health, property, and specialty lines — shared entity graph, per-LoB models, cross-line ring detection from day one.

Up to 4 LoB model stacks
Shared claimant + repairer graph
Similarity over full claim archive
Bi-weekly model reviews
Regulator-ready SAR workflow
Enquire for pricing
Fleet
Enquirefor pricing
Network-wide consortium

Consortium deployment across multiple carriers with shared ring-detection infrastructure, dedicated fraud engineering, and sovereign data handling per jurisdiction.

Unlimited LoBs + carriers
Dedicated fraud engineering team
Consortium graph + shared signals
24/7 monitoring + on-call
Jurisdictional deployment
Talk to us

FAQ

Common questions.

Don't see your question here?

Ask us directly

Glossary

The vocabulary behind every fraud verdict.

A quick reference for the terms that show up in claims-fraud detection — the language your SIU, compliance, and regulator all work in.

Soft fraud
Opportunistic exaggeration

Padding of an otherwise legitimate claim — inflated invoice, added damages, stretched narrative. Typically handled by adjustment rather than outright denial.

Hard fraud
Organised / intentional

A deliberately fabricated or staged claim — often coordinated across multiple claimants and providers. Targeted for blocking and regulatory reporting, not negotiation.

Staged accident
Engineered motor-fraud event

An incident orchestrated to generate a claim — often involving scripted collisions, phantom passengers, and collusive repairers feeding inflated invoices.

Bust-out
Premium-then-claim pattern

Policyholder pays premium briefly, then stages a large claim early in the cycle — a pattern detectable through tenure + claim-size + graph features.

Inflated claim
Artificially increased payout

Legitimate incident with damage or medical billings padded beyond actual loss. The canonical soft-fraud signature that document-pattern models target.

Collusion network
Linked-party conspiracy

Claimants, repairers, clinics, and sometimes adjusters acting together. Network-graph analytics are designed to see these as clusters, not isolated cases.

SIU
Special Investigation Unit

The carrier's internal fraud-investigation team. SIUs consume the flagged-claim queue, run interviews, and decide on block, adjust, or SAR filing.

SAR
Suspicious Activity Report

A regulatory filing raised when fraud or money-laundering is suspected. Required in many jurisdictions under FATF-aligned AML rules, and central to carrier compliance.

EXIF metadata
Exchangeable image file data

The metadata every digital photo carries — device, timestamp, geo. Cross-checking EXIF against the claim narrative is a first-line image-authenticity test.

Image tampering
Pixel-level manipulation

Any edit to a photo after capture — splicing, cloning, re-save, or AI generation. Detected via pixel-noise residual, error-level analysis, and device attribution.

Pixel forensics
Compute vision for authenticity

Techniques that surface manipulation invisible to the human eye — error-level analysis, noise residuals, and CFA-pattern checks on every claim image.

Champion-challenger
Parallel model operation

Running a new model (challenger) alongside the live model (champion) on real traffic to validate lift before promotion. Standard in fraud-model deployment.

False-positive rate
Incorrect flags / total flags

The proportion of flagged claims that turn out to be legitimate. Core trade-off: lower FPR means fewer customer frictions but more missed fraud.

Chargeback cycle
Post-payout dispute window

The period after a claim is paid during which the carrier can still recoup funds on discovered fraud. Shrinking this cycle is a key ROI driver for pre-payout detection.

Explainable · SIU-ready

Your fraud stack, engineered.

30-minute scoping with a senior engineer and an SIU-systems operator. You'll leave with a model plan, integration sketch, and realistic timeline — not a sales pitch.