AI-Inspect - an artificial intelligence tool for instant inspection of commercial properties

ended

Tensorflight has created AI-Inspect, an artificial intelligence tool for the instant, remote and fully automated analysis of the different features of commercial buildings for the insurance and reinsurance sector. We develop the most advanced computer vision methods based on deep learning and are the first company that analyses not only satellite/aerial images but also street-level pictures. Due to the high scalability of our innovation, we are able to quickly provide information on large portfolios of commercial properties from any part of the world. We will cover a market niche where at the moment no one can provide a high-quality, rapid service and facilitate the credible risk assessment of properties such as retail and manufacturing sites, office and multifamily buildings, hospitals, schools, etc. AI-Insect will help insurance and reinsurance companies to reduce costs of on-site surveys and reduce losses caused by improper valuation. As a result, both the owners and the users of the properties will share lower costs of the insurance coverage. In a 5 years perspective, we estimate to analyse no less than 15.5 million queries at a cost of approx. 1 EUR each. Better insurance coverage in the real estate sector is extremely important bearing in mind the climate change. Accurate risk assessment is crucial for large insurance and reinsurance companies to step into more risky markets in the developing countries. The ultimate result is a globally shared risk of catastrophic events and higher stability of the economy in different regions of the world. In this feasibility study, we plan to perform a comprehensive market analysis that will refine the pricing strategy to fit regional specificity and identify new partners to conduct pilot testing on new markets. A technical assessment will also be conducted to explore innovative solutions that will allow us to boost AI-Inspect's capabilities, keeping us far ahead of the competition.