يقول الأشخاص الذين يعملون في مجال تأمين المخاطر إن الذكاء الاصطناعي ينطوي على مخاطر كبيرة للغاية بحيث لا يمكن التأمين عليه.

## Insuring the Uninsurable: When Risk Assessors Flinch at AI

The adage “if anyone can quantify risk, it’s the insurance industry” faces its ultimate test with artificial intelligence. So, when the very professionals whose livelihoods depend on meticulously assessing, pricing, and mitigating risk declare AI “too risky to insure,” it sends a powerful tremor through the burgeoning tech landscape.

This isn’t merely a cautious stance; it’s a profound statement on the unprecedented challenges AI presents. Insurers thrive on historical data, actuarial tables, and predictable patterns. AI, however, introduces a realm of novel uncertainties: opaque decision-making (“black box” algorithms), rapid and unpredictable evolution, potential for widespread systemic failures, and a murky landscape of liability when an autonomous system causes harm.

The inability to accurately model catastrophic scenarios, assign clear accountability, or even understand the full scope of potential misuse leaves underwriters in uncharted territory. Their hesitation underscores a critical reality: while AI promises transformative benefits, its inherent complexities and the unknown unknowns make it an outlier even for those accustomed to gambling on the future. It’s a stark reminder that innovation, no matter how promising, must grapple with the fundamental questions of risk, responsibility, and the limits of our ability to predict the next big thing.