The legal tech group connected to UK-based law firm Kennedys has launched what it’s calling the ‘first fully explainable neuro-symbolic AI risk analysis solution’. It’s called Kennedys IQ SmartRisk, plus it’s ‘devastatingly trustworthy’ they say.
The firm explained that the tool will ‘transform how insurers approach policy review, liability, and coverage analysis by accelerating review and decision-making whilst improving accuracy and consistency’.
The product, which comes out of the Kennedys IQ group, sounds intriguing….but what on Earth is neuro-symbolic AI?
Here’s how they put things:
‘Unlike pure GenAI solutions that rely solely on probabilistic outputs, Kennedys IQ SmartRisk leverages a hybrid approach, integrating Large Language Models (LLMs) with insurance knowledge modelled using Evidential Reasoning (ER), and Belief Rule Base (BRB) methodologies. Kennedys IQ SmartRisk goes beyond pure language processing, providing insurers with a structured decision-making framework.
‘This innovative approach ensures fully auditable and explainable analysis, overcoming industry concerns around AI’s ‘black-box’ opacity. The system does not require large amounts of upfront data and can be configured for any line of business and integrated into client systems in just a few weeks.’
In short, there is a high level of rules-based programming going on here too, and it’s not all probabilistic ‘black-box’ output that you can’t ever really get into the inner details of.
OK, still interesting. Who will want this?
It’s designed for claims professionals, brokers, and underwriters across global markets, including the London Market, specialty and general insurance.
It will help to support ‘automated risk assessments and policy analyses, enhance claims governance through structured decision-making, extract critical data from complex documents – including handwritten notes – and improves compliance and accuracy, while reducing inconsistencies in decision-making’, they added.
Karim Derrick, Chief Product Officer, Kennedys IQ, said: ‘There is a growing demand for AI-driven automation in insurance, particularly within specialty and complex risk underwriting. Policy wording analysis and claims handling are complex, inconsistent, and prone to human error, and GenAI chatbots today lack auditability, making them unsuitable for insurance risk assessment.
‘Our interest is helping clients identify, manage and mitigate risk through the SmartRisk tool which can fill this gap to ensure transparency in decision making and regulatory compliance. In doing so, it eliminates the ‘black box’ concern around other AI models, while improving efficiency, minimising human error and providing insurers and brokers with deeper data-driven insights. Kennedys IQ’s SmartRisk frees up insurance professionals to focus on high-value decision-making while maintaining full control over risk analysis.’
The move makes a lot of sense – in fact, all of Kennedys’ moves into developing tech solutions for the insurance world do. Why? Because insurers have to deal with huge volumes of data – both in document form and otherwise. They’re also quite cost-averse given the volumes of policies being written, and the claims they have in motion at any time.
So, tech that can help with this, e.g. policy analysis, or risk assessment, is going to help reduce costs and push up their margins.
Note: as explored previously by AL, this is a great example of thinking about ‘end goals’ and client value, rather than just looking inward in terms of how AI can help the law firm by itself. Please see AL article here for more: ‘What Is AI For? We Need To Think About End Goals’.
The idea is illustrated here in the chart and the meme (which aptly features J.F. Kennedy.)
And,
Neil Mody, a US partner at the firm, added: ‘The SmartRisk U.S. model is designed to help insurance adjusters navigate complex regulatory frameworks, including New York Insurance Law 3420. By mirroring the proprietary coverage evaluation process honed by seasoned professionals over decades, SmartRisk enables adjusters to reliably pinpoint key coverage issues, and make well-informed decisions.’
And Richard West, Head of Client Innovation and Director of Kennedys IQ, concluded: ‘SmartRisk is devastatingly trustworthy and our proprietary response to the AI revolution. Built on our deep professional insights, it seamlessly integrates specialist judgement with cutting edge technology. It is transparent, explicable, and designed to protect our clients’ and their customers’ reputations. We encapsulate professional integrity at the heart of Kennedys IQ SmartRisk.’
More here.