Insurance sector risks 'AI washing' as spending outpaces actual deployment
Ninety-nine percent of insurers have launched generative AI initiatives, yet fewer than half have deployed them into active business functions. Even fewer-15 percent-report measurable returns on investment.
This gap between AI spending and tangible results has created what industry experts now call "AI washing": the misrepresentation or overstated claims about AI's effectiveness in insurance products and services.
A report released in April 2026 by Simplifai analyzed data from McKinsey, EY, Deloitte and Swiss Re. It found that while 83 percent of carriers spend more than $5 million annually on AI initiatives, the actual business impact remains unclear. Only 15 percent of insurers saw measurable improvements in combined ratio, cycle time or loss ratio.
The pressure comes from above, not customers
The drive to adopt AI is not customer-led. Michael James, technology consulting director at Simplify Consulting, said customers are actually concerned about insurers using AI. The real pressure comes from competitors, boards, investors and fear of falling behind.
"Much of the current market noise is better described as AI gloss than AI strategy," James said.
Claud Bilbao, vice president for underwriting and distribution at Cowbell, described much of the AI washing in insurance as "dressed up" rule-based automation. These systems don't improve understanding of a client's risk profile or provide proactive guidance.
The FCA is increasingly scrutinizing false or exaggerated technology claims, Bilbao added. As more firms incorporate AI, the risk grows that AI washing will distort how insurance is provided, managed and governed across the sector.
Governance frameworks lag behind adoption
The speed of AI adoption has outpaced the development of governance frameworks. Thomas Barrett, partner at law firm Weightmans, said businesses risk accumulating latent liability because adoption has been prioritized over robust oversight.
An EY-Parthenon survey of UK financial services leaders found that only 8 percent believed current AI frameworks provided clear guidance-compared to 19 percent globally.
Many AI systems are adopted with "relatively limited scrutiny," Barrett explained. Boards lack the technical expertise to interrogate them properly, or supplier claims are accepted without question. This is particularly concerning given the potential for bias and opaque decision-making.
"Businesses are deploying AI before governance, compliance and testing frameworks are sufficiently matured," Barrett said. "There is a real risk that some organizations are already accumulating liabilities that simply have not surfaced yet."
The liability problem is murky
Somesh Mukherjee, vice president for innovation, data and AI at Acord Group, warned against treating AI governance as an "abstract compliance exercise." The industry is moving from high-level ethics concerns into an era where technical accountability will matter most.
If an AI agent handles submissions, prices risk or triages claims, organizations must ensure absolute model traceability. If a model drifts or its data lineage is unclear, the organization won't survive a modern regulatory audit.
"You can buy the software, but you cannot buy your way out of the risk," Mukherjee said.
Bilbao said the UK's sector-by-sector approach to AI regulation leaves small and medium enterprises exposed to a "massive expectation gap" between what AI promises and what it delivers. A poorly implemented system that makes up facts or acts on biased data leaves the business completely exposed.
That exposure quickly becomes a professional indemnity claim, a directors and officers claim for failing to govern operational risk, or a regulatory fine for misrepresentation. Where the liability sits remains "worryingly murky" because there aren't yet enough test cases.
Brokers need to help clients understand they own the liability for the AI tools they deploy, Bilbao said.
Focus on measurable operational gains
James stressed that the focus should shift to applying and governing AI responsibly to turn it into a measurable operational advantage. Zurich UK formalized its commitment to ethical AI adoption by signing the Community Union's Fair Future for Finance Responsible AI Charter in May 2026.
Asha Kumar, chief technology officer at Zurich UK, said boards and senior leaders need stronger understanding of AI-specific risks to embed them into governance and operating models. Risk is not purely technical-it's driven by how AI is used in real-world contexts.
Leadership must strengthen oversight through a balanced approach across technology, process and people, supported by training, structured risk assessments and governance forums that bring together business owners and risk experts.
Elizabeth Wooliston, chief of markets at Artificial, said digitalization at board level means staying involved beyond initial approval. Organizations that get value from digital change are clear about what they're trying to improve and remain engaged as initiatives move from design into delivery.
Operational efficiency offers clearer returns
The strongest near-term AI opportunity for insurers lies in operational efficiency, not flashy consumer-facing applications. Claims handling, service delivery, case triage, engineering productivity and customer acquisition all offer more immediate and measurable potential.
Some of the most underused opportunities sit in "relatively unglamorous areas" like structured data, clearer decision frameworks and better integration between existing systems. These don't attract headlines but make an immediate difference to how consistently underwriting intent translates into outcomes.
James warned that underwriting "often attracts the loudest AI claims" but should attract the greatest caution. Regulatory attention in the UK remains focused on consumer understanding, claims handling and service quality, making disciplined operational improvement more valuable than broad AI theater.
Learn more about AI for Insurance and Generative AI and LLM implementation challenges.
Your membership also unlocks:
