The market in one read
Open Insurance Brazil for insurers is, in one read, a shift from gathering data to interpreting it. The SUSEP-led open insurance regime standardizes how consumer insurance data is shared, with free, informed and prior consent, across participating companies, and the moment that standardized data arrives, the bottleneck stops being access and becomes the ability to turn shared data into a decision. That is the strategic line for any carrier in Seguros e Danos (P&C): the rails are being built by the regulator, and the advantage moves to the intelligence layer that sits on top of them.
State of the P&C insurance market
Brazil's Seguros e Danos (P&C) market grows double digits per year, and that pace, not any single premium figure, is what sets up the tension for underwriting. Company structure has not kept up with the acceleration. Submissions arrive faster than teams, intake processes and core systems were built to absorb, which is the gap Open Insurance starts to address from the data side.
The regime itself is a SUSEP construct. SUSEP, the Superintendência de Seguros Privados, supervises the open insurance system through a layered governance structure, first established in 2021 and made permanent at the end of 2024. Its stated aims, according to SUSEP, are to bring innovation to the sector, promote competition, and improve products and services for consumers. Participation is mandatory for some companies and voluntary for others, and CNseg sector statistics track the underlying P&C premium volumes the regime acts upon. The headline for insurers is not the size of the market in any one report. It is that a standardized, consented data layer is being assembled across the sector, in phases.
What is pressuring underwriting
The first pressure is arithmetic. Double-digit premium growth outpaces operational structure, so the volume of submissions widens faster than the capacity to assess them. Standardized, consented data offers a way to enrich and pre-fill submissions without simply adding headcount.
The second pressure comes from distribution. Brazilian distribution is broker-led, and the corretor (broker) routes business to the insurer that quotes fastest and most consistently. According to Capgemini, 60%+ of brokers choose an insurer by response speed. Standardized data that arrives pre-structured shortens the path from submission to quote, which is a direct conversion lever on the distribution side.
The third pressure is fragmentation. A single risk is still assessed from scattered inputs. PDFs, broker e-mails, registration data spread across systems, and prior-policy history the insurer often cannot see. Gartner estimates that companies lose 20-30% of their time organizing unstructured data, and that is precisely the manual reconciliation Open Insurance reduces by putting registration and contract-usage records into common, machine-readable formats. There is also a structural reason carriers stall here. BCG finds that 70% of insurers do not execute on innovation because of IT limitations, which is why an approach that does not touch the core matters.
Consent is the unlock that ties these together. Public data flows without consent, but personal data sharing requires consent that SUSEP defines as free, informed and prior, expressed by electronic means. Consent is both the legal gate and the practical trigger for the richer data underwriting actually needs.
Risk, fraud, and the AI shift
Open Insurance changes subscrição (underwriting) from a data-gathering exercise into a data-interpretation one, and that is where AI and Machine Learning enter. When a consented consumer's registration data and prior-policy history arrive in standardized form, the underwriting journey changes at specific points rather than all at once.
Document reading is the first. Standardized fields cut the volume of free-form documents a team has to parse, and for the documents that remain, intelligent document reading with ML extracts the fields and maps them onto the same schema. Risk scoring is the second. With richer structured inputs, a scoring layer can rank submissions against the insurer's risk appetite and underwriting manual, separating what sits inside appetite from what needs a human underwriter. This is the step where shared data becomes a decision and not just a record. Pricing is the third. Standardized history and usage data support more granular premium (prêmio) calculation where the line of business allows it.
Fraud detection improves on the same inputs. When prior-policy and claim history is visible across participants, it is harder to hide adverse history at quotation, and ML models can flag inconsistencies between declared and shared data earlier in the journey rather than at claim time. Quality improves too, because standardized fields mean the same data carries the same meaning across insurers, which lowers reconciliation error.
The constraint on all of this is regulatory, and it is real. Open Insurance personal data is consented personal data under LGPD, and automated decisions that affect the consumer carry transparency and review expectations. Models may only use data within the scope and duration the consumer authorized. SUSEP supervision combined with LGPD pushes toward decisions that can be explained and audited after the fact, so any AI scoring layer has to log which data drove a decision, the model version, and the appetite rules applied. Explainable and auditable is not a feature here. It is a condition of operating.
Where WIR fits
WIR is the external AI layer that turns Open Insurance data into explainable, auditable underwriting decisions, on top of the systems the insurer already runs, never in their place. WIR does not build the Open Insurance rails and does not replace the core. It consumes the shared, standardized data as one more input into its pipeline, sitting 100% external to the insurer's IT, with no core migration. WIR is not an insurer, a broker, or an MGA, and it does not carry risk. It automates the quotation and underwriting journey according to the insurer's own risk-acceptance policy.
In practice, the open-insurance data feeds the same flow WIR runs on any submission. Intelligent document reading extracts and standardizes fields. Broker enrichment adds score, conversion history and prioritization, cross-referencing external context. A risk and fraud engine, built as a multi-factor ML model calibrated to the insurer's risk appetite and underwriting manual, returns a risk score and an automated decision. Dynamic pricing calculates a risk-adjusted premium. The final step issues a quote, an automatic decline, or an escalation to a human, always with an explanation, writing back to the policy core and returning the audit trail. Two products carry this. Underwriter Intelligence automates the quotation journey so underwriters spend their time on risk and business development, and Smart Sales scores upsell and next-best-action across the portfolio so penetration and retention move together.
Every step runs encrypted and LGPD compliant, and every decision returns a full audit trail, which is what makes the open-insurance use case viable under SUSEP supervision. WIR's only public traction at this stage is a POC in execution with a global insurer in the Transport line. The positioning holds across all of it: WIR is the AI layer for insurance, on top of existing systems, calibrated to each insurer's appetite.
Outlook
The regime should keep phasing from public data toward fully consented personal data and transactional services, and the strategic value to insurers rises as the personal-data and services phases mature, because that is when underwriting-grade data actually flows. Competitive pressure on response speed will tend to reward carriers that can ingest standardized data and quote faster through the broker channel, turning data access into a distribution advantage rather than a compliance task.
The binding constraint is unlikely to be data availability. It is the ability to interpret shared data against appetite, in real time, in a way that stays explainable and auditable, and to do it on top of existing core systems rather than through a multi-year re-platforming. Where that leaves Brazilian insurers is fairly clear. Open-insurance data is becoming a shared input layer, and the differentiator moves to the intelligence layer that consumes it. An external AI layer that reads documents, scores risk against the underwriting manual, flags fraud, and supports pricing, while preserving auditability under SUSEP and LGPD, is the practical way to capture the upside without rebuilding the core.
Frequently asked questions
What is Open Insurance Brazil and who regulates it?
Open Insurance Brazil is the regime that standardizes and shares insurance sector data in phases, regulated by SUSEP. SUSEP, the Superintendência de Seguros Privados, supervises the system through a layered governance structure, first established in 2021 and made permanent at the end of 2024. Public data flows openly, while personal data sharing requires consent that SUSEP defines as free, informed and prior. Participation is mandatory for some companies and voluntary for others.
How does shared data change insurance underwriting?
Shared data turns underwriting from a data-gathering exercise into a data-interpretation one. When a consented consumer's registration data and prior-policy history arrive in standardized form, teams parse fewer free-form documents, scoring layers can rank submissions against the insurer's risk appetite, and pricing can use more granular history. Fraud detection improves too, because prior-policy and claim history visible across participants makes adverse history harder to hide at quotation.
Does WIR consume Open Insurance data to enrich submissions?
Yes. WIR consumes the shared, consented Open Insurance data to enrich submissions and scoring, as an external AI layer on top of the insurer's core. The standardized data feeds the same pipeline WIR runs on any submission: intelligent document reading, broker enrichment, a risk and fraud engine calibrated to appetite, and dynamic pricing. WIR does not build the Open Insurance rails. It treats the consented data as one more input.
Is Open Insurance data handled in line with LGPD?
Yes. Open Insurance personal data is consented personal data under LGPD, and WIR runs every step encrypted and LGPD compliant. Models may only use data within the scope and duration the consumer authorized, and automated decisions affecting the consumer carry transparency and review expectations. Every WIR decision returns a full audit trail logging which data drove it, the model version, and the appetite rules applied, which is what makes the use case viable under SUSEP supervision.
Does WIR replace the insurer's core to use Open Insurance?
No. WIR does not replace the core. It is the external AI layer, 100% external to the insurer's IT, with no core migration. WIR consumes Open Insurance data as one more input into its pipeline and writes decisions back to the policy core the insurer already runs. WIR is not an insurer, a broker, or an MGA, and does not carry risk. It automates the quotation and underwriting journey per the insurer's own risk-acceptance policy.