InsureVision, a provider of video-based risk intelligence for insurers, focused on applying advanced AI to underwriting and pricing, has announced the launch of two new products designed to help insurers return to profitable underwriting.
The company has introduced TeslaStick for personal lines embedded insurance and FleetVision™ for commercial fleet risk assessment, targeting what it describes as a multi-bn dollar challenge facing the insurance industry, according to InsurTech Insights.
The launch comes as insurers continue to struggle with loss ratios driven by poor risk selection, limited behavioural insight and rising claim severity.
Traditional underwriting models often rely on blunt mechanical or historical indicators that fail to capture real-world driving context. InsureVision positions its new products as a response to this problem, enabling carriers to access richer, real-time risk signals without requiring costly new hardware or complex integrations.
InsureVision develops AI-driven technology that analyses video data to generate predictive risk intelligence for insurers.
By applying advanced scene understanding techniques similar to those used in autonomous driving systems, the company aims to give underwriters deeper insight into driver behaviour, road conditions and situational context. This approach allows insurers to move beyond proxy metrics and towards more accurate, forward-looking assessments of risk.
TeslaStick is designed to turn existing in-vehicle camera infrastructure into an underwriting tool for personal lines insurance. Tesla vehicles already feature forward-facing cameras that continuously record footage via the Tesla Dashcam. TeslaStick connects through the vehicle’s glovebox USB port, enabling access to this continuous video stream. A companion smartphone app detects potential risk events using a custom risk Transformer Network, retrieves the relevant footage and sends it to InsureVision’s AI for contextual analysis.
The product is distributed via a B2B2C model, allowing insurers to white-label TeslaStick for Tesla-owning customers. This enables personalised, risk-based pricing without insurers needing to build or maintain their own AI capabilities. The solution supports embedded insurance strategies by leveraging hardware that is already installed in vehicles.
FleetVision™, meanwhile, is aimed at commercial fleet insurers grappling with the challenge of accurately pricing fleet risk and identifying high-risk drivers early. While fleets already collect large volumes of dashcam footage, this data has typically been used only after incidents occur. FleetVision analyses existing footage to detect behavioural indicators such as following distance, situational awareness and lane discipline, helping insurers predict crash risk before claims arise.
According to the company, FleetVision allows carriers to price fleet risk accurately from inception and intervene proactively with drivers who may trigger high-cost claims. By focusing on contextual scene understanding, the product seeks to replace crude mechanical proxies with more meaningful risk signals.
“The future of all insurance is embedded,” says InsureVision CEO Mark Miller. “Look at what Lemonade’s doing with Tesla insurance – they aim to leverage how much Full Self-Driving you use, which is a proxy at best. We’re going across everything. Whether you’re using autopilot or driving manually, we’re getting a truer picture of risk by understanding the complete driving environment and the context, not just the metrics. We’re simply accessing the Tesla Dashcam feed through their USB port. The phone detects braking events, fetches the relevant footage, and our transformer AI analyses the context. It’s that straightforward.”
Keep up with all the latest FinTech news here
Copyright © 2026 FinTech Global
