Bernstein Research has identified 10 best practices for investors integrating generative AI into asset management, emphasizing the importance of structured data, targeted workflows, and measurable outcomes. The brokerage emphasizes the need for clean, structured internal data, prompt engineering, centralizing data, breaking down daily workflows into specific use cases, and developing vertical AI champions. To scale AI effectively, firms should prioritize top use cases, structure offsite exercises, and engage in regular knowledge-sharing sessions within teams. Developing vertical AI champions in departments such as equities, fixed income, legal, or compliance ensures solutions remain close to real use cases. Dedicated AI talent is also needed, with some firms assigning specific team members to focus on AI tools or hiring external specialists. Tools like Daloopa and ModelML are cited for model automation and internal data integration. Early engagement with implementation partners or adopting ready-made AI tools can speed progress without requiring deep technical expertise. In the future, organizations should prepare to work with hybrid teams of humans and artificial intelligence, requiring robust data infrastructure and governance. Clear metrics to evaluate Gen AI’s impact include operational efficiency, error rate reduction, time to generate insights, volume of AI-generated ideas, and comparisons with human output. Success in portfolio management can be assessed by time saved during scenario analysis and the frequency of bias avoidance in decisions post-AI implementation.