About 10 percent of the property and casualty insurance industry’s incurred losses and loss adjustment expenses each year are fraudulent[i] During the five-year period from 2013 to 2017, the fraud amounted to approximately $30 billion each year.[ii] The most common fraud is when an insured pads or inflates actual claims, misrepresents facts on an insurance application, or submitting claims for damage or injuries which never occurred. Most insurers use software technology to detect and combat the fraud.
One of the most common technologies used today are automated red flags, which automatically highlights suspected fraudulent activity. The software can look at the claim and compare it to similar situations or find patterns of claims which do not look normal. Sophisticated data analysis can sometimes uncover relationships between groups across multiple claims which might represent organized fraud rings. Insurers realize no single technology is sufficient to combat opportunistic or organized fraud. A relatively new technology they are evaluating is voice AI.
Nemesysco Ltd., a privately held company founded in 2000 with headquarters in Israel, has developed technology which uses voice analysis for emotion detection, personality, and risk assessment.[iii] McCormick explains how a Slovakian insurer is using the technology,
Insurer Allianz-SP Slovakia, a subsidiary of Allianz ALIZF +2.66% group, handles claims using Nemesysco’s voice-stress analysis technology. The tool picks up people’s reactions to a set of scripted questions asked by the claims handler. The system looks for a combination of markers, such as tiny pauses when a person is speaking, that may indicate the speaker is providing false information, according to Allianz-SP Slovakia. “The aim is to pay a claim without any problems immediately and to prevent any fraud-like exaggeration of a claim,” says Jaroslava Zemanová, head of control and special activities at Allianz-SP Slovakia.[iv]
Allianz-SP Slovakia emphasizes the voice analysis is not proof of wrongdoing. The company views the analysis as a first stage in detecting the possibility of fraud. An investigation would determine if there is more evidence which may justify rejecting a claim. The company says the technology is saving it time and money so far. [v]
[i] “Background On: Insurance Fraud,” Insurance Information Institute (2018), https://www.iii.org/article/background-on-insurance-fraud
[ii] Ibid.
[iii] “Investigation Focus Tool,” Nemesysco (2019), http://nemesysco.com/
[iv] McCormick, “What Ai Can Tell from Listening to You”.
[v] Ibid.
