The article also highlights the importance of good data for successful AI implementation, with 39% of respondents to the Gartner survey citing a lack of data as a top barrier. Companies are advised to focus on a limited set of data with potential for reuse. Other challenges include respecting data use agreements and dealing with vulnerable populations. The need for a centralized approach to AI across the company, governance, security, and demonstrating real ROI are also emphasized. Despite these challenges, companies are encouraged not to be paralyzed by them, but to start with something that works and shows value, and build from there.
Key takeaways:
- Companies are interested in generative AI, but face challenges in implementation due to technical complexity and a lack of skilled personnel.
- Studies show that only a small percentage of companies have successfully implemented generative AI projects at scale, and even fewer have seen a positive impact on earnings.
- Data readiness is a significant part of AI readiness, with a lack of data being a top barrier to successful AI implementation. Companies are advised to focus on data that can be reused across multiple use cases.
- Companies need to balance the potential benefits of generative AI with the need for governance, security, and demonstrable ROI. A centralized approach to AI across the company is recommended, along with reusing successful elements to increase delivery speed.