Risk Detection

Detect what matters. Out of the box.

Isomer comes pre-trained with specialized agents designed to catch the most expensive risks across claims and underwriting. No custom training required.

Cover all cases
Legal & Bad Faith
Medical & Injury
Fraud & Abuse
Complex & Emerging
Financial & Coverage
Compliance
Legal & Bad Faith

Stowers / Time-Limit Demands

Identifies settlement demands with strict deadlines that, if missed, create bad faith exposure above policy limits.

Example

"Incoming letter stating "Demand for policy limits of $50,000" with a "expires in 14 days" clause or citing Stowers v. American Indemnity."

Action: Prevent Bad Faith setups; protect policy limits.
Legal & Bad Faith

Attorney Representation

Detects the first signal of legal counsel to prevent improper direct contact with the insured.

Example

"Email from a domain like @forthepeople.com, signature block containing "Esq." or "Attorney at Law," or a formal "Letter of Representation" attachment."

Action: Stop direct contact; route to senior counsel.
Medical & Injury

MIST / Medical Creep (Fraud)

Flags "Minor Impact Soft Tissue" claims where medical treatment disproportionately exceeds vehicle damage.

Example

"A repair estimate for <$800 combined with medical bills for "chiropractic adjustment" exceeding $5,000 or 30+ visits."

Action: SIU Referral: Catch $30k+ medicals on <$2k damage.
Medical & Injury

TBI / "Invisible Injury" Creep

Monitors medical records for subtle keywords suggesting high-value Traumatic Brain Injuries not initially diagnosed.

Example

"Medical notes mentioning "dizziness," "post-concussive syndrome," "DTI scan" (Diffusion Tensor Imaging), or "neuro-cognitive testing" after a minor accident."

Action: Flag "DTI Scans" to prevent reserve shocks.
Compliance

Sanctions / OFAC Compliance

Real-time screening of all payee and claimant names against federal sanctions lists.

Example

"A claimant or vendor name matches an entry on the SDN (Specially Designated Nationals) list during intake or before payment issuance."

Action: Real-time federal SDN list screening.
Legal & Bad Faith

"Bad Faith" Setup Phrasing

Detects adversarial language specifically crafted to stage a future bad faith lawsuit against the carrier.

Example

"Correspondence stating "unreasonable delay," "failure to settle," "ignoring our offer," or "putting financial interests ahead of the insured.""

Action: Flag "Ignoring offer" traps beyond policy limits.
Legal & Bad Faith

Evidence Preservation

Identifies demands to retain physical or digital evidence to prevent spoliation sanctions.

Example

"A letter demanding preservation of dashcam footage, black box/ECM data, surveillance video, or the vehicle itself for inspection."

Action: Secure dashcam, ELD, or black-box data now.
Fraud & Abuse

Provider/Attorney "Clustering"

Visualizes network relationships to flag suspicious pairings of law firms and medical providers.

Example

"Detection that Attorney Smith has referred 15 different claimants to Dr. Jones' Pain Clinic in the last 6 months (a "mill" pattern)."

Action: Uncover collusive fraud mills across claims.
Financial & Coverage

Litigation Funding (TPLF)

Detects involvement of Third-Party Litigation Funders, which often inflates settlement demands.

Example

"Correspondence from entities like Oasis Financial, US Claims, or Cartiga, or liens filed by non-medical financial institutions."

Action: Detect hedge-fund backed lawsuits early.
Financial & Coverage

Aggregate Limit Erosion

Alerts when cumulative payouts across multiple claims threaten coverage limits.

Example

"Multiple claims against a single policy or account approaching the aggregate limit for the policy period."

Action: Reinsurance Alert: Flag hits to geography limits.
Complex & Emerging

Construction: Latent Defect

Identifies complex construction defect allegations that may trigger multi-year policy analyses.

Example

"Complaint alleges "water intrusion" or "structural failure" occurring over 5 years, triggering a "continuous trigger" coverage analysis."

Action: Multi-policy continuous trigger analysis.
Complex & Emerging

Professional Liability: E&O

Flags potential Errors & Omissions claims, often signaled by missed deadlines or administrative failures.

Example

"Email from a client alleging "failure to file" or "missed statute of limitations" regarding a professional service provided by the insured."

Action: Admission check for "missed deadline" triggers.
Complex & Emerging

Environmental: PFAS / AFFF

Scans for specific chemical contaminants (forever chemicals) that represent systemic mass-tort exposure.

Example

"Complaint or FNOL mentions "firefighting foam," "groundwater contamination," "PFAS," or "PFOA" exposure."

Action: Early warning of systemic chemical mass-torts.
Fraud & Abuse

Multi-Claimant Linkage

Detects when multiple unrelated claims stem from the same root cause or event (Class Action signal).

Example

"5 separate claims filed in one week all alleging "food poisoning" at the same restaurant location or "product failure" of the same device."

Action: Class Action Alert: Detect systemic cause of loss.
Fraud & Abuse

AI-Generated Evidence

Uses metadata analysis to flag potentially manipulated photos or documents (Deepfakes).

Example

"A photo of vehicle damage is flagged because metadata indicates it was created by "Adobe Photoshop" or an AI generation tool, not a camera."

Action: Metadata/Deepfake photo checks for fraud.
Legal & Bad Faith

Punitive Damage "Setup"

Identifies allegations of "gross negligence" intended to bypass policy limits or exclusions.

Example

"Complaint uses specific phrases like "malicious intent," "conscious disregard for safety," or "corporate negligence" rather than just standard negligence."

Action: Detect "Corporate Negligence" document demands.
Financial & Coverage

Policy Gap / Silent Exposure

Flags claims that may fall into coverage gaps or "Silent Cyber" exposures not intended to be covered.

Example

"A General Liability claim alleging "data breach" or "ransomware loss" which should be routed to a Cyber policy (or excluded)."

Action: Identify risks improperly classed (Silent Cyber).
Fraud & Abuse

Social Media "Sentiment"

Monitors external platforms for viral negative sentiment or public complaints about a claim.

Example

"A claimant posts a video of the accident or a complaint about the adjuster on TikTok, X (Twitter), or Reddit that gains significant traction."

Action: Flag claims trending on TikTok/X or news.
Legal & Bad Faith

"Reptile Theory" Tactics

Detects plaintiff strategies focused on "safety rules" and "community danger" to inflate jury verdicts.

Example

"Deposition questions or letters asking: "Do you agree safety is a top priority?" or "Did the defendant violate safety rules that protect the community?""

Action: Detect safety rule/community danger language.
Legal & Bad Faith

High-Risk Jurisdiction / Venue

Alerts when a lawsuit is filed in a jurisdiction known for excessive plaintiff verdicts.

Example

"Suit filed in Cook County, IL, Philadelphia, PA, or South Florida, triggering an immediate "Removal Analysis" to try and move to Federal Court."

Action: Removal Analysis: Move to Federal court.
Medical & Injury

Subrogation Indicators

Detects early mentions of third-party fault that suggest an opportunity to recover costs.

Example

"FNOL or adjuster notes mention "other driver ran red light," "defective tires," or "contractor error" caused the loss."

Action: Detection of product or third-party part failure.
Medical & Injury

Medical Severity Escalation

Flags casual mentions of medical developments that will significantly impact reserves.

Example

"An informal email update mentioning "scheduled for surgery," "hospital admission," or "permanent impairment" that hasn't been formally coded."

Action: ICU/Surgery keywords for reserve accuracy.
Legal & Bad Faith

Statute of Limitations (SOL)

Monitors the countdown to legal filing deadlines to prevent missed defenses or default judgments.

Example

"Alert generated: "SOL expires in 30 days" for a specific claim based on the Date of Loss and State Jurisdiction."

Action: Deadline Watch: Detect "statute running" misses.
Financial & Coverage

Excess Carrier Notification

Automatically notifies excess/umbrella carriers when reserves breach reporting thresholds.

Example

"Primary policy erosion hits 50%, triggering an auto-notice to the Excess Carrier to prevent late reporting denials."

Action: Notify over-limit layers at 50% hit.
Legal & Bad Faith

Critical Flagging

A catch-all for high-severity events requiring immediate senior intervention.

Example

"Inbound mail contains words like "Fatality," "Sexual Abuse," "Class Action," or "Media Inquiry.""

Action: Immediate escalation for high-severity events.
Load More

Underwriting Risk Detection

The same signal library extends to underwriting. Detect portfolio risk across submissions and active policies before it becomes claims exposure.

Classification Verification

Systematic verification of risk types, industry codes, and coverage categories across your active portfolio.

Emerging Risk Scanning

Continuous monitoring for emerging categories — PFAS, latent construction defects, cyber exposure in non-cyber lines — across submissions and renewals.

Portfolio Pattern Detection

Cross-portfolio pattern identification revealing concentration risk, aggregate exposure, and correlated losses invisible in any single policy.

Partner With Us on New Signals

We're actively working with carriers to develop new signal detectors. If your organization has proprietary risk patterns, litigation data, or emerging exposure categories you want detected, we'd like to build them with you.

Ready to deploy these signals?

Isomer connects to your communications in minutes and starts flagging these risks immediately.