Most conversations about AI in insurance underwriting start with speed. But Peeyush Rai, Founder and CEO of Weav.ai, argues that focusing only on velocity - faster quotes, more straight-through processing - ignores what makes commercial underwriting different. "If you look at underwriting, especially in commercial lines and specialty lines, it's very different from personal lines," he said. "Each risk is very different, and you need a broader lens to assess the risk. It's not just a number thrown by an AI model."
That distinction shapes Weav.ai's approach. The company has helped carriers cut quote turnaround times sharply, but Rai warns that speed without discipline is dangerous. "You could be turning around lots of quotes, but if they are not profitable it's not a good thing. You're doing bad things faster." So the objective is two-fold: faster decisions and better risk selection. The first place to apply that thinking is before an underwriter even touches a submission.
Cleaning up the submission mess
In commercial insurance, underwriting teams spend a huge amount of time just gathering, validating, and cleaning submission data. Brokers and agents send incomplete packages, information arrives in inconsistent formats, and carriers often outsource parts of this prep work. Rai describes it as a "huge upfront overhead." Weav.ai automates much of that pre-underwriting process, extracting data from documents, cross-checking entries, and flagging missing information.
The result: an 85% reduction in manual effort for submission review. The system handles most extraction and validation, but it doesn't operate in a vacuum. A confidence-based model sends only uncertain cases to a human. "The 15% that goes to human-in-the-loop are the cases where AI says, 'I'm not really confident about this particular data point or information or document that has come in,'" Rai said. For some carriers, underwriting preparation that once took three or four days now takes 15 to 20 minutes.
Filtering noise before it reaches the desk
Speed also helps solve a subtler problem: submission noise. Commercial carriers often receive large volumes of business that doesn't fit their appetite, burying the high-quality opportunities under less relevant applications. "Some brokers and agents spray carriers with lots of submissions for the same business," Rai said. By issuing indicative pricing much earlier in the process, Weav.ai lets brokers and agents quickly see whether a risk is likely to be competitive. That filters out unsuitable opportunities, letting underwriters focus on better risks.
For insurers exploring AI for Insurance, the key is that automation doesn't just accelerate - it reshapes workflow. Carriers can dedicate more time to complex risks while routine, clearly in-appetite submissions move faster.
Modeling risk the way a carrier thinks
Once a submission reaches the underwriting stage, the platform shifts to decision support. Instead of a flat list of variables, Weav.ai builds a structured representation of the business, its operations, and exposures. The system researches business operations, validates classifications, and constructs a view that mirrors how the individual carrier evaluates risk. "Two property carriers will not look at the risk the same way," Rai said. Even within the same line, underwriting philosophies, product structures, and risk appetites differ.
The platform uses AI scorecards that assess risk before the underwriter sees it. Carriers set thresholds for automatic approval, decline, or referral. This means routine risks can flow through without human intervention, while complex cases still get underwriter attention.
Building trust and connecting decisions to outcomes
Rai compares the adoption of automated underwriting decisions to self-driving cars. "Initially, we are cautious of what the car is doing, but then you get used to its behaviour. From there, you start trusting it for certain things." The same gradual trust-building applies to AI-driven underwriting. Metrics back up the progress: pre-underwriting manual effort drops 85%, and initial quote turnaround goes from days to 15 minutes.
Underlying all of this is data, but not just more data. Rai said the challenge is making sense of it. Weav.ai combines internal carrier data, third-party information, and publicly available sources into knowledge graphs that create a single, structured view of the risk. "The key is to combine them into one view of the risk," he said. "Pricing and rating engines are basically input and output. If you give them high-quality input, you'll get a high-quality output."
The company is also building systems that connect underwriting decisions with policy performance over time. Audits, endorsements, renewals, and claims activity flow back into future decision-making. "Risk is never static," Rai said. Tracking that dynamism helps underwriters and product managers see trends at the portfolio level, not just point-in-time.
Why this matters for insurance
Commercial underwriting is fundamentally a judgment-intensive process across diverse risks. AI that only accelerates quoting can amplify bad decisions. The opportunity lies in tools that reduce submission noise, mirror a carrier's specific risk appetite, and tie underwriting decisions to long-term portfolio performance. For insurers, the shift from speed-alone metrics to accuracy and consistency changes how they deploy underwriters - letting them focus on the risks that truly demand human expertise.
Your membership also unlocks:
