5 Q’s for Jakub Dryjas, CEO of Tensorflight
The Center for Data Innovation spoke with Jakub Dryjas, CEO of Tensorflight, a Polish-U.S. company that combines satellite, aerial, and ground-level imagery with artificial intelligence (AI) technologies to aid property risk assessments. Dryjas spoke about Tensorflight’s partnership to help rebuild Ukraine since the onset of the war.
Gillian Diebold: What is property intelligence, and what are its benefits?
Jakub Dryjas: Collecting accurate data on properties has always been challenging. In the past, property owners and insurers were required to conduct onsite inspections to assess the state of a specific asset. As technology progressed, insurers started leveraging public data, previous property appraisals, and databases built on public record aggregations. But it’s easy to see how these processes and databases can be tainted with incomplete, out-of-date, or even false information. When data is incorrect, it increases the risk that insurers and underwriters create policies and premiums that don’t truly reflect the value of the property.
In the insurance industry, underwriters can use property intelligence to underwrite policies with more accuracy and create more competitive insurance products. Property intelligence involves providing more real-time, accurate data for insurers to act upon and incorporating modern solutions, like Tensorflight, to provide that data in an easily accessible way; for instance, directly via API or a web portal.
Diebold: Tensorflight works in the “insurtech” space. What does that mean, and what kind of data is collected for this type of technology?
Dryjas: Insurtech refers to the technologies helping to improve the efficiency of the insurance industry. While technology has transformed many aspects of the insurance industry, the pace of change has been slow, and we’ve yet to see much meaningful impact on property underwriting. Many insurers still rely on public records and decades-old surveys to collect information on properties and inform their decisions. As a result, too many buildings end up underinsured, highlighting the risks posed by manual, time-consuming property inspection and valuation processes.
We believe that the best way to help insurers minimize risk and improve their processes is with more up-to-date, relevant data collection and analysis. We are particularly focused on geospatial data, which can include spatial data, ground-level imagery, aerial survey imagery, high-resolution satellite imagery, and multispectral imagery.
By providing insurers with more robust geospatial data and ensuring that the data is current, underwriters can more accurately estimate the risk of insuring property. For example, Tensorflight can provide detailed data on a property’s facade material and condition, construction type, replacement cost, occupancy type, and roof geometry. Insurers can use this data to reduce premium leakage, provide customers with instant and automated policy quotes, improve their loss ratios, significantly decrease onsite inspection costs, and handle claims more efficiently.
Diebold: Your company is helping with the rebuilding of Ukraine in partnership with the Kyiv School of Economics. Can you talk a little bit about Tensorflight’s contributions to that project?
Dryjas: Since the onset of the Ukraine war in February 2022, the KSE Institute, a leading think tank and the analytical unit of the Kyiv School of Economics, has been spearheading an initiative to collect, evaluate, analyze, and document information on the damage inflicted to infrastructure resulting from the Russian invasion.
The KSE Institute sought out external expertise to provide intelligence on a larger scale and, since April 2022, Tensorflight has been providing and analyzing satellite imagery to assess the damage caused to infrastructure, specifically within Bucha, Irpin, and Mariupol, on a pro-bono basis. Through analyzing satellite imagery and rooftop geocoding information, we can offer insight into the extent of the damage caused to any particular building in the analyzed areas, while also providing the exact geographic coordinates of the buildings.
Our team of AI experts trained a deep neural network model to detect and classify war-damaged buildings based on satellite images, which is achieving accuracy rates of over 90 percent. Our next plan is to analyze satellite imagery covering Lysychansk, Melitopol, and Popasna. However, satellite imagery is extremely costly. We are facing upwards of €5 million to scan the 100,000 square kilometers of Ukraine still awaiting damage assessment. Additionally, this data becomes out-of-date quickly when further damage is inflicted, so we are currently looking for sponsors and partners to help finance and continue the project.
Diebold: How can your product be of value to individual consumers in addition to businesses?
Dryjas: When property insurance businesses suffer from a lack of reliable data, it affects the quality of the policies and services they can deliver to their end customers or policyholders. Tensorflight helps insurers create more accurate and detailed assessments of properties, which enables them to offer customers more tailored and effective insurance policies. Furthermore, by eliminating the need for manual inspections, insurers can generate more accurate quotes in less time, which improves the customer’s experience and, ultimately, retention and satisfaction.
Without this accurate and up-to-date information about the properties being insured, insurers may be unable to properly assess the risks involved and offer appropriate coverage. This can lead to policies that are inadequate or overly expensive and leave policyholders vulnerable to financial losses in the event of a claim. Our main goal is to reduce risk for both insurers and policyholders, resulting in a more competitive and sustainable property insurance market for everyone.
Diebold: What differentiates Tensorflight from other organizations in the property intelligence space?
Dryjas: Tensorflight combines ground-level, aerial, and the highest commercial resolution satellite imagery available on the market to generate the most accurate property data. This data ranges from information about roof conditions to damage analysis and appraisal following natural disasters and within conflict zones. In these situations, on-the-ground inspections are not possible, and other existing sources of imaging do not provide the level of detail needed for accurate assessments. Our platform can also combine recent and past data, allowing insurers to see changes to properties from as little as six months ago, as well as how properties have changed over time, empowering them to make more educated and informed decisions.
We are a leader in the safe, ethical use of property data, and our global platform features comprehensive coverage of millions of properties across North America, Europe, Great Britain, Australia, and Southeast Asia.
https://datainnovation.org/2023/03/5-qs-for-jakub-dryjas-ceo-of-tensorflight/
