In this industry report, Ours shared expert insights on the strategic implementation of generative AI in organizations.
He emphasized that successful GenAI adoption requires a methodical approach. “You need to critically evaluate what you need, where you’re deploying it, and what benefit you want from it,” he said. “You need to measure that productivity and ensure you’re getting that value as you make the investment.”
Ours advised organizations to focus on business-level problems rather than individual ones, such as improving efficiency, enhancing customer experiences, or driving innovation, to ensure solutions scale effectively across the enterprise. Aligning ROI with business needs and truly understanding what makes a project successful are also critical.
“If your job is to sell lemonade, your strategy is not AI,” he said. “AI can help you achieve your strategy, it can make your supply procurement more efficient, and it can make your marketing more effective — but it can’t sell lemonade for you, so it shouldn’t be your strategy.”
On the technical side, Ours stressed the importance of quality data integration, noting that “when you start connecting data sources, you’re setting up and anchoring the large language model in latent space. And so, if those data sources aren’t the right ones or they’re erroneous, they’ll lead the model astray and produce suboptimal results.”
He also discussed assessing project feasibility, learning new skills, and best practices for moving your project forward.
While GenAI is changing processes and helping organizations realize value, simply implementing it isn’t enough.
“Large language models are like very smart, very capable, very powerful interns — but they have no business awareness, no applicability, no understanding outside what you tell them to do and how to do it,” he concluded. “And so if you don’t provide them with that business understanding, data, and context, they can do quite a bit of wrong.”