What impact do you think DigiLog's Advanced Validation Solutions (AVS) – incorporating Voice Risk Analysis (VRA) technology will have on society?
Do you think Voice Risk Analysis (VRA) can make a serious dent in the £14 billion annual fraud bill?
Is Voice Risk Analysis technology a “lie detector”?
How does Voice Risk Analysis work?
Isn't DigiLog just a tool for detecting Insurance Fraud?
Do you think VRA has been subjected to enough testing and scientific research?
Do you think customers should be made aware beforehand that they are being subject to Voice Risk Analysis?
Do you think insurers should be allowed to refuse a claim on the back of VRA results?
What other methods of combating claims fraud might be more socially acceptable by consumers?
Could VRA lead eventually to a reduction in insurance premiums?
Do you think it could be perceived as an infringement of civil liberties?
 

What impact do you think DigiLog's Advanced Validation Solutions (AVS) – incorporating Voice Risk Analysis (VRA) technology will have on society?

The potential is nothing short of immense. Assessing risk in what people say to us is probably something human beings have been doing since the first human posed the very first question! By and large, we have usually relied upon our untrained, highly subjective gut feelings and instincts to steer our trust in what we are told – and often with catastrophic results. We have been easily fooled for far too long – but that need not be the case anymore. DigiLog's AVS represents a quantum leap in identifying risk in conversations, replacing our grossly imperfect subjectivity with two highly intelligent processes working in tandem and within a logical framework. From identifying risk in job applicants, especially those applying for sensitive positions in education, health care/ social welfare or security, through those that fraudulently apply for, or claim on, State Benefits/ Banking/ Credit/ Insurance products and ultimately to empowering the police when interviewing suspects or witnesses.

Do you think Voice Risk Analysis (VRA) can make a serious dent in the £14 billion annual fraud bill?

Yes – a huge dent – but only where people are defrauded as a result of a conversation and only when using it as part of DigiLog's AVS solutions. Whilst VRA can greatly enhance the ability to spot signs of truth or deception, it is only when deployed alongside our Advanced Questioning Techniques and Narrative Integrity Analysis skills that it helps tip the balance hugely against the fraudster. Our dual approach ensures that we ask the right questions, obtain the detail necessary to work with and maximise both the positive effects on the genuine and the negative effects on the deceptive.  We are already having a very significant impact in Insurance and know that we have just as much to offer in Banking, the Public Sector Benefits system and Criminal investigations.

Is Voice Risk Analysis technology a “lie detector”?

It must be understood that lying is not a unified set of feelings. Lying occurs as a result of a deep, conscious and logical decision making process that manifests as a result of a person's intention to deceive. One person may lie to protect themselves from harm, while another may lie to gain profit, or even just to make a joke. There is no fixed set of characteristics (physiological or psychological) that differentiate lies from truth. However, VRA uses a patented and unique technology to analyse the emotional structure of what is being said, deploying a wide range spectrum analysis of micro-changes in the speech waveform itself (not micro tremors) through which we can learn about the relationship between emotion, stress and cognitive activity. For example, measurable, anomalous relationships occur when there is conflict between what the brain knows to be true but what it is instructing the subject to say. This is particularly true when the subject has a price to pay were the truth discovered. When combined with the powerful questioning techniques and NIA skills within DigiLog's AVS, we open up an incredible window into truth and risk of deception.

How does Voice Risk Analysis work?

The unique Voice Risk Analysis technology utilised within our process doesn't just measure stress, it seeks to interpret the complete emotional structure evident in individually assessed conversations (excitement, pain, fear, cognitive activity and more). The correlation between these various emotional stress characteristics enables us to determine whether a statement was, amongst other things, truthful, deceptive, uncertain and inaccurate. It does this in every sentence your subject says and independently analyses each one against baseline measurements taken in a short phase of anodyne questions that individually calibrate the software at the beginning of the conversation.

Isn't DigiLog just a tool for detecting Insurance Fraud?

DigiLog's AVS process has a multitude of professional applications. It is designed to assess risk in any conversation, whether over the telephone, face-to-face or captured in a recording. The diversity of applications includes:

• Pre-employment screening, especially for sensitive positions such as in education, health care/ social welfare or security/ law enforcement
   
• Screening conversations that relate to applications for, or claims on, State Benefits/ Banking/ Credit/ Insurance products
   
• Empowering the Police when interviewing suspects or witnesses

Additionally, we firmly believe that organisations should be considering DigiLog AVS not just as a means to detect deception, but as a tool with which the genuine can be quickly identified and either fast-tracked towards their legitimate goal (such as an insurance claim settlement) or eliminated from the potential intrusion and inconvenience of an investigation. 

Do you think VRA has been subjected to enough testing and scientific research?
 
The Nemesysco technologies deployed as part of DigiLog’s AVS process, have been subjected to a number of studies that show them to be highly effective in detecting both truth and deception. However, all of the available research focuses only on the VRA technology - none examine DigiLog AVS's dual approach, i.e. using VRA in tandem with our powerful techniques (e.g Advanced Questioning, Conversation Management and Behavioural Analysis skills), often critical in achieving results. To research this area robustly is extremely difficult, as AVS is not a perfect science. Every scenario we analyse contains a multitude of factors that make it different from any other; every conversation is different and, of course, each human being applying the AVS process will perform differently.

Consequently, the best research material on AVS comes from a tremendously important source: the results achieved by, and the testimonials made by, our customers - many of whom are highly respected, blue chip corporations. These results come from real-time, real-world, operational deployments. Many insurers, some using our solution since 2002, have seen huge improvements in their fraud identification rates. Highway plc and many other insurers, some using our solution since 2002, have seen huge improvements in their fraud identification rates. One has saved more than £11.5 million from the processing of only motor 19,000 claims. Provident Insurance in the UK is currently defeating an 29% of all their unfiltered claims - saving more than 25% of their claims reserves in the process.

Do you think customers should be made aware beforehand that they are being subject to Voice Risk Analysis?
 
For much of the past, customers have been largely unaware of what formed part of any process instigated to validate their claim or application. Additionally, we have only previously been able to take at best an educated guess as to whether a claim was suspicious or not, resulting in the possibility of both the genuine and the fraudsters being subjected to scrutiny. Indeed some of the more traditional methods, including profiling, financial status checking, and surveillance, could be considered to be in the grey zone when it comes to whether or not it is ethical to use them at all.
 
Recent regulation and guidance in a variety of areas has assisted in removing some of that greyness. DigiLog recommends as a minimum that companies inform customers who phone in to their call centres that all telephone calls relating to applications and claims may be tape recorded and analysed for the purposes of fraud prevention and detection.
 
We recommend that companies are quite open about using the VRA technique as part of an overall process designed to sift the truthful from the disingenuous. After all, the concept is not new. Human beings have been using their own computer – i.e. their brain, to assess truth or deception in conversations since we first learned to communicate. Publicising the use of VRA and associated AVS techniques, does two jobs – reinforcing the stance against premium inflating fraudsters for the benefit of the honest, whilst acting as a deterrent to those that are thinking of cheating.

Do you think insurers should be allowed to refuse a claim on the back of VRA results?
 
On the basis of VRA alone? No – that is not until VRA analysis are proven to be accurate ‘beyond reasonable doubt', as per the burden of proof in criminal cases. That may happen in the future - but we are not quite there at the moment. Human beings remain owners of the decision making process – albeit they are now highly trained by DigiLog and highly focused by the variety of techniques embraced by AVS. Their decisions are made on a case by case basis and according to the evidence available.

What other methods of combating claims fraud might be more socially acceptable by consumers?
 
Since early 2002, some of our clients have publicly announced their use of DigiLog, attracting a great deal of media attention world-wide.  Despite this, we have yet to see any evidence at all that suggests VRA as being socially unacceptable to consumers. In fact quite the reverse. Our clients have received highly positive feed back from their customers. Even the 10% of customers that the ABI suggests are fraudsters seem to have quietly accepted that their days are numbered – withdrawing or accepting declination of their claims without fuss.

Ultimately, the test of social acceptability depends on may factors – the test population itself, the size of the problem, the proportionality of response, etc. Currently companies can end up using toffee hammers to try and crack mountains or sledgehammers to crack peanuts. That simply isn't good enough. DigiLog has moved its clients on to a point where they are now capable of professionally deploying the right tools in the right areas and in a cost effective fashion. It is enormously beneficial to do this and doubly so when, at the same time, service levels to genuine customers are so dramatically enhanced. This is precisely what DigiLog's AVS Solutions achieve – so only those at the margins should have any reason to object.

Could VRA lead eventually to a reduction in insurance premiums?
 
In short, it is the right combination of methodologies and technology that is now hauling investigation and validation out of the dark ages and into the 21st Century. VRA very definitely is already playing a major role in this - whether it is in helping to quickly identify the genuine, quickly spotting the deceptive or both. Either way, we are very confident that solutions incorporating VRA and DigiLog's other AVS elements will contribute to lower costs and begin to stabilise the annual premium hikes and ultimately deliver the margin that will subsequently drive them down.