In 2023, funding for AI start-ups soared to nearly $50 billion, reflecting the growing interest in AI’s potential. OpenAI CEO Sam Altman envisions AI as a “super-competent colleague” that knows everything about your life, though some find this prospect unsettling. However, as the value of tech companies plummets, there’s a real question about whether the AI bubble has already burst. We’ve seen boom and bust cycles before, and now we’re left to wonder: Is AI the “garlic bread” that will enhance our work-life balance?

Hannah Fry, in her book Hello World, makes a compelling case about AI’s in-built biases. Feeding AI a dataset of previous decisions, such as in sentencing, often leads to biased outcomes rather than a utopian, unbiased approach. The old adage, “garbage in, garbage out,” remains true. Fry also offers a “magic” test for assessing AI claims: replace all technical jargon with the word “magic.” If the sentence still makes sense, it’s likely nonsense.

AI does hold promise, particularly in fields like physics and biology, where pattern recognition can be life-saving. For instance, AI could revolutionize the analysis of MRI scans or cancer prediction. Businesses are also being pitched AI solutions that promise increased profitability and efficiency. Yet, the first step often involves paying a hefty sum just to find out how they can help—a proposition that invites scepticism.

Over the past 25 years, online insurance purchases have grown, especially for personal insurance and micro-businesses. Yet, many business owners still prefer the personalized service of brokers, despite the millions invested in technology by insurers. Complex business needs simply can’t be met by a one-size-fits-all approach.

At Vista, we’ve handled over 500 deals in the past decade, from tech start-ups to multinationals. Doing the job right requires extensive reading and analysis. AI could indeed streamline insurance due diligence, offering increased efficiency, improved accuracy, and enhanced decision-making. As entrepreneurs and brokers, we recognize the need to adapt to stay competitive, but we’re also aware of the challenges that AI brings.

Key Opportunities and Challenges in AI Integration:

1. Streamlining Data Collection and Analysis: AI can make data gathering more efficient, but it’s only as good as the data it’s trained on. Poor data quality can lead to flawed outputs and poor decision-making.
2. Enhancing Risk Assessment: AI can offer deeper insights into risks, but overreliance on technology could lead us to neglect our own expertise. AI should support, not replace, human judgment.
3. Automating Routine Tasks: Robotic Process Automation (RPA) could reduce errors and improve consistency, but system failures could disrupt operations.
4. Improving Fraud Detection: AI can spot potential fraud, but false positives and negatives remain a concern.
5. Enhancing Decision-Making: AI can provide data-driven insights, but bias in AI models and the need for human oversight are critical issues.

Conclusion

AI has the potential to transform insurance due diligence, from data collection to risk assessment and decision-making. However, we’re not there yet, and the hype must be approached with caution. At Vista, we’re investing in AI’s early stages, using it where it can enhance our services, but we recognize that human expertise and intuitive analysis are irreplaceable, especially in the complex world of Private Equity deals.

Garlic bread? Not quite yet.