How Multimodal AI Could - Retool Global Crisis Response - Ben Ellencweig, Jessica Lamb, Jon Spaner, and Mihir Mysore,

Based on historical, predicted, and real-time data points, gen AI can create and analyze thousands of scenarios to assess the probabilities of various crises and their impacts across a range of severities, conduct risk assessments, and help authorities prepare appropriate response strategies. [...] And it’s important to verify the reliability of the data to ensure the simulations’ outcomes are as accurate as possible.18 It is also important to address the potential risks of content generated by gen AI, including the risk that images, videos, and text will be weaponized to spread climate disinformation and panic. [...] The ability to track, understand, and explain the origins of gen AI algorithms and models fosters trust and allows both emergency managers and the community at-large to engage with system development. [...] Early engagement, communication, and transparency can build trust and ensure that the use of AI technologies aligns with the needs and values of the communities these tools are meant to serve. [...] These interventions can help responders coordinate efforts to mitigate the impact of polycrises and enable quicker and more resilient community recovery.21 About the Authors Ben Ellencweig is a senior partner in McKinsey’s Stamford office and global co-leader of QuantumBlack, AI by McKinsey; Jessica Lamb is a partner in the New York office and the healthcare and public sector leader of QuantumBlac.