Two to four yes/no questions, a real-time analysis, no AI model in the traditional sense, and yet an accuracy rate that’s causing a stir among insurers. Manjit Rana, Executive Vice President of Insurance at Clearspeed, explains in an interview withthebrokernews how neuroscience-based technology is revolutionizing the fight against fraud, and why the Swiss market has long been more than just an insider tip for Clearspeed.
On 9 June 2026, Manjit Rana was a guest at the “Reinsurance Rendez-vous Zurich,” organised by the Swiss InsurTech Hub in collaboration with the Boston Consulting Group. What he presented there sounded, at first glance, like science fiction: a technology originally developed for military special forces and now used in day-to-day insurance operations to detect fraud based on the human voice without biasbiases. We asked him about it.
Manjit, Clearspeed has its roots in the military sector. What does a technology developed for special forces have to do with the day-to-day work of insurance brokers and claims handling teams?
The underlying challenge is the same: screening large numbers of people quickly and fairly, without relying on background or profile. In a military setting, that means vetting recruits who may gain access to sensitive intelligence or weapons. In insurance, it means assessing claimants or applicants at speed and without bias. In both cases, Clearspeed helps determine who or what to trust through voice, rather than by inferring risk from demographic or historical data.
Clearspeed unlike other predictive models that leverage AI, Clearspeed is grounded in neuroscience. To outsiders, that sounds like a significant difference, but why exactly? What makes this approach fundamentally different?
Insurers already use AI to sharpen fraud models, using larger datasets to build tighter segments based on age, location, historical behaviour or claims timing. Clearspeed takes a different approach: The signal comes entirely from the voice.
The signal itself: when someone answers a question where something is genuinely at stake, the brain reacts involuntarily, producing measurable changes in the voice. Clearspeed reads those changes, not the words spoken, but a signal. This technology is grounded in decades of neuroscience research.
How it’s captured: a short set of automated yes/no questions, delivered by phone or online, with no human interviewer. That removes interviewer bias and ensures every person is assessed identically.
What sets it apart: most tools look backward, using background checks, credit scores, or pattern-matching against records. Clearspeed provides an in-the-moment signal that cannot be coached, profiled, or falsified.
Key properties: the signal sits below the level of language, so it’s language agnostic and is also agnostic to gender, race, and location. Risk is flagged at the level of each individual question, not just the overall session, giving teams precision rather than a blunt score.
The output: a result returns within seconds, minimal risk or elevated risk. Clearspeed does not make the decision; it tells your team exactly where to focus.
You describe your technology yourself as a ‘metal detector’: not a judgement, but an early filter. How do you ensure that this metaphor delivers what it promises in practice?
A metal detector doesn’t decide whether a passenger flies; it signals whether metal is present, so security staff know who needs a closer look. Clearspeed works the same way. It signals whether a risk indicator is present in response to questions tailored to the specific risk being assessed, whether that’s a claim, an application, or an onboarding decision. It never determines the outcome. It tells the organisation, with a high degree of accuracy, whether a case can be fast-tracked or needs to follow the usual human review. Because the questions are built around the specific risk in question, the same approach adapts to virtually any use case across the insurance value chain.
A 20- to 30-fold return on investment is a strong claim. Can you illustrate this with a specific example from everyday insurance practice, and what changes for a claims handler working with Clearspeed?
The return isn’t purely about catching more fraud, that’s the common misconception. It comes from three sources at once. First, fraud detection itself: high-risk claims are flagged early and investigated properly, stopping claims that would otherwise be paid simply to avoid the cost of a dispute. Second, operational efficiency: when the majority of claims clear as low-risk within seconds, skilled investigators spend their time only on the minority that genuinely need them, lowering the cost per fraud detection substantially. Third, and often overlooked, is speed of settlement for genuine customers. Allianz UK saw a 59 per cent lift in immediate claims settlement alongside a 36 per cent lift in average fraud-loss savings per claim. As Allianz stated: fraud solutions don’t need to disrupt the customer journey, they can improve it. The 20 to 30-fold average return across our client base reflects the compounding effect of all three factors, set against a low unit cost for the technology itself.
Clearspeed is used for claims management, underwriting and onboarding. Where does the greatest leverage lie today, and where do you see untapped potential?
Today, the greatest leverage lies in claims (FNOL) and fraud detection, where most of our insurance clients began, with growing extension into underwriting and point-of-sale. The untapped potential extends further across the value chain: identifying employees, including remote call-centre staff, who may be capturing and selling claims data to disreputable third parties; and detecting cases where an applicant or intermediary withholds risk-relevant information to secure a cheaper premium. Wherever an organisation needs a fast, unbiased answer to whether a risk signal is present, there is a role for Clearspeed.
Partnerships such as those with Allianz or Zego show that Clearspeed is already well established in the UK market. What is required technically, organisationally and culturally for an insurer or reinsurer to really get off the ground?
Clearspeed is well established across the UK and US, partnering with the majority of major P&C insurers in the UK, with growing deployment across Europe, Asia, and Latin America. Technically, the barrier to adoption is low: the solution runs through a web browser, so an insurer can be live and deriving value without IT resources or integration into policy or claims systems. Deeper integration adds further value as usage scales, but it isn’t a prerequisite to start.
Culturally, very little needs to change. Applicants or claimants who clear the automated questionnaire can be fast-tracked to settlement or onboarding, improving the experience for genuine customers. Investigation teams gain the confidence to focus their time on the cases that flag elevated risk, and can spend more time and empathy with the customers who need it. Both effects reinforce each other from day one.
At the ‘Reinsurance Rendezvous Zurich’, you were a guest of the Swiss Insurtech Hub. How do you perceive the Swiss market, and what distinguishes it from the UK market or other EMEA markets in which you operate?
The Swiss and UK markets are both mature and well regulated, but structurally distinct. Nineteen of Switzerland’s 26 cantons run monopoly cantonal insurers for building and natural-hazard cover, and compulsory, community-rated health insurance (LAMal) has no UK equivalent. Switzerland also punches above its weight in reinsurance, with Zurich the world’s third-largest hub. Distribution is relationship-driven, with around 1,000 FINMA-registered brokers serving a stickier, lower-churn market than the UK’s price-comparison-led, increasingly digital-direct model.
The contrast sharpens around fraud. The UK reports hard detected figures, £1.16 billion in 2024 from roughly 98,400 fraudulent claims and 684,800 blocked applications, while Switzerland estimates around one in ten claims payouts involves fraud, a higher claimed prevalence despite the UK catching more in absolute terms. The difference is infrastructural: the UK has a mandatory, centralised claims record, an industry fraud bureau, and a dedicated police unit; Swiss data-protection law sets a high bar on cross-insurer data-sharing, and the industry’s shared tool covers only the eight largest carriers. This is exactly where Clearspeed adds the most value: because our signal requires no external or personal data, it lets insurers focus investigation resources precisely where the risk is real, regardless of how fragmented the surrounding data infrastructure may be.
Clearspeed completed a Series D funding round of 60 million dollars in 2025. What will this capital be used for, and what role does the DACH region play in your growth strategy?
The raise funds international growth across our two core markets: Government, Defence & Security, and Enterprise, the latter weighted heavily toward insurance and banking. DACH, and Switzerland specifically, is a clear priority within that expansion, given the market’s sophistication and its distinct fraud and regulatory profile.
Fraud is becoming increasingly sophisticated; deepfakes and AI-generated voices are no longer the stuff of science fiction. How does Clearspeed stay one step ahead of a threat landscape that is itself evolving with the help of AI?
Long procurement cycles mean insurers are almost always working with fraud-detection tools that lag the sophistication available to a determined bad actor. Clearspeed sidesteps the AI arms race by design: an individual knows if they have manipulated evidence, and Clearspeed sits before the submitted evidence is even examined. If no risk signal is detected, a claim can be fast-tracked without the cost of sourcing and reviewing evidence. If a signal is detected, the investigation team can apply the full range of tools, including expert review, to the claim and its evidence. This is particularly effective for exaggerated property claims, where the claimant genuinely owns the item and holds a valid receipt but knows the item hasn’t actually been lost or stolen.
Finally: if you could give a Swiss insurance broker, who is hearing about Clearspeed for the first time, a single piece of advice, what would it be?
Innovative solutions are rarely built specifically for insurance. More often, they’re adapted from thinking and technology proven against similar challenges elsewhere. Clearspeed was built to assess risk in high security and military screening, and has since been deployed across government, probation services, sports integrity, insider threat, banking, and insurance, worldwide. The organisations that benefit most are the ones willing to look outside their own industry for a solution already proven elsewhere.
My advice: stay open to trying it. Every significant shift in an industry looks unnecessary right up until it becomes standard practice. The value is best understood in a conversation, not a brochure, and that’s a conversation we’d welcome.
Binci Heeb asked the questions.
Manjit Rana has been shaping the UK insurance industry for nearly four decades. From early broker software and reporting platforms to telematics, the rise of insurtech, and the fight against AI-driven fraud schemes, he has consistently been ahead of the curve. As EVP of Insurance at Clearspeed, he works with insurers to address emerging challenges in fraud, claims, and customer trust through innovative technology solutions. He is a globally recognized thought leader in insurance innovation, an entrepreneur, and a conference speaker with deep expertise across insurance markets in the UK, the US, and APAC. Manjit has mentored founders, advised boards, and serves as an Entrepreneur-in-Residence at the University of Nottingham. Most recently, he was and was named Insurance Personality of the Year at the 2026 British Insurance Awards.
See also: AI in Reinsurance: Between Hype and Measurable Added Value
Search:
Sponsors:
