In a recent analysis, McKinsey & Company, a management consulting firm, examines how artificial intelligence (AI) could influence the future direction of the global insurance sector, arguing that insurers, distributors and technology providers that prepare early may be better positioned as industry structures evolve.
According to McKinsey & Company, the global insurance industry has experienced steady premium growth over the past two decades, but improvements in operating leverage have been limited across property and casualty (P&C), life and health insurance.
The firm estimates that gross written premiums have increased by around 4.9% annually since 2005, reaching approximately $8.3 trillion in 2025, while profits before tax have grown by around 4.3% during the same period, reaching approximately $580 billion. McKinsey notes that rising capital requirements have contributed to this slower profit growth.
The company states that the insurance sector has historically been resistant to major disruption. While developments such as globalisation, digitalisation and platform-based business models have reshaped many industries, McKinsey says they have had a more limited impact on insurance’s underlying economic structure.
Competitive positions have shifted gradually, capital movement across regions and business lines has remained relatively slow, and public markets have generally continued to view insurance as a stable industry with predictable earnings.
McKinsey & Company notes that private capital has introduced innovation in areas such as balance sheet management and investment strategies, but has had less impact across other parts of the insurance value chain. The firm suggests that the insurance industry of 2026 would remain recognisable to executives who viewed the sector in 2006.
The company acknowledges that this stability has also brought benefits. The insurance industry has continued to provide significant shareholder returns through dividends and share buybacks, while maintaining an important role in supporting economies and societies, including during major events such as the global pandemic.
However, McKinsey argues that the sector now faces increasing pressure from artificial intelligence, which has the potential to influence four long-standing industry challenges: slower growth and declining relevance, high distribution costs, limited productivity gains and a historically gradual pace of change.
McKinsey identifies a growing gap between rising global risks and the insurance industry’s ability to provide coverage. The firm states that insurance revenues have grown more slowly than many major industries and global GDP, with personal lines representing 1% of global GDP in 2023 compared with 1.2% in 2019. According to the company, growth in developed markets has often been driven by pricing increases rather than expansion into new areas of risk.
The firm highlights significant protection gaps in emerging risk areas. McKinsey estimates that the global natural catastrophe protection gap reached $133 billion in 2025 and notes that less than 1 per cent of global cyber costs are currently insured, representing a potential gap of around $900 billion. The company argues that insurance is becoming less aligned with an increasingly complex risk environment.
McKinsey & Company suggests that AI could create opportunities for the sector through new risks, new solutions and expanded market access. The company notes that AI may introduce additional areas of insurable risk, including AI liability, non-physical business interruption and workforce-related risks linked to AI adoption.
It also suggests that technologies such as parametric insurance, embedded micro-coverage and real-time data-driven policies could make insurance accessible to customers and risks that have previously been difficult to serve economically.
The company also highlights a potential shift from traditional risk transfer towards broader risk partnerships. McKinsey explains that conventional insurance has largely focused on responding after losses occur, whereas AI could enable insurers to provide continuous monitoring, insights and prevention support before incidents happen.
According to McKinsey, examples of this approach could include telematics systems that provide real-time driving guidance while adjusting premiums, commercial risk management supported by satellite and Internet of Things data, and AI-enabled health support designed to improve health outcomes. The company states that these approaches already exist in limited areas but have not yet become central to the wider insurance proposition.
McKinsey further argues that AI could improve access to insurance markets by strengthening underwriting and claims capabilities. The company notes that some emerging risks, including climate-related property risks, cyber threats and AI-related exposures, remain difficult to price because insurers lack sufficient reliable data and predictive confidence.
The firm suggests that improved data analysis, continuous model updates and more accurate claims assessment could help insurers price risk more effectively and expand coverage. McKinsey states that carriers able to develop these capabilities early may gain advantages through improved loss ratios, stronger pricing confidence and the ability to enter markets where competitors may remain cautious.
However, McKinsey also highlights potential challenges. The company notes that some digital risks may not behave like traditional insurance exposures, as shared infrastructure, interconnected supply chains and common technology dependencies could create highly correlated losses. According to McKinsey, insurers entering these areas will need strong analytical capabilities to understand how risks develop and spread rather than simply creating new products.
The company concludes that AI is likely to create both opportunities and challenges for insurance organisations. McKinsey states that success will depend not only on adopting AI models, but on using the technology to create distinctive capabilities and long-term competitive advantages.
