“Longer term, I think people haven’t fully understood the power of generative AI beyond their own personal use cases. But now, the central theme is really testing out where it can go.” – Nick Heinzmann, Head of Research at Zip
Most procurement teams are likely in some stage of testing, adopting, or optimizing their use of new AI technologies. Those that aren’t are probably thinking about it and wondering how to start. Clearly, AI is a road we are all heading down together – even if some are further ‘ahead’ than others. When it comes to deciding exactly when, where, and how to integrate AI into the procurement tech stack, teams need a solid understanding of the best approaches and the risks to look out for along the way.
In this AOP podcast episode, Kelly Barner and I speak with Lance Younger, CEO of ProcureTech, and Nick Heinzmann, Head of Research at Zip, about current best practices around adopting AI tools that will maximize value creation while not exposing the business to unnecessary risk.
Here, in their own words, are highlights from this timely conversation and the key takeaways Lance and Nick had for our listeners:
Understanding Difference Between ‘AI’ and ‘Generative AI’
Lance: “At a basic level, AI allows you to analyze and work with historic data sets and allows you to be predictive. Generative AI allows itself to be able to model by itself, create new outputs, create different outputs. Hence, we’ve had some instances of organizations saying, ‘It hallucinates.’ It’s actually just being super creative!”
Trends in Procurement AI Adoption
Lance: “Over the last six months, especially as we were heading into 2024, we have seen a lot more AI adoption. It’s been around specific use cases and people effectively running pilots, running activities, and creating their own private LRMs and the like. We have also seen a lot of the more leading digital solutions advance and embed generative AI into their solution sets. Many of them were able to solve one or two problems that they have not been able to solve before, which makes it a lot more exciting.”
The Role AI Plays in Procurement Orchestration
Nick: “If you look at it this way, orchestration started as a new layer. It sits on top of everything you might call an ‘automation layer.’ There are S2P tools or suites that sit underneath. Orchestration as a value prop provides central control, seamless integration, and visibility into everything that sits underneath.
In my view, AI – especially generative AI – is an accelerant of what orchestration is already doing because, from that top layer, you can embed AI down into any current S2P process without making you necessarily have to adopt a specialized AI within those specific tools or modules.”
Most Common Procurement Testing Spots for AI
Lance: “The question here is spot on: ‘Is generative AI going to be used in RFPs and contract drafting?’ If you are not doing that already, then you are behind – 100 percent. Whether it is in the generation of a contract, the review of a contract, or the creation of an RFP, or the identification of suppliers – AI is being applied in all those different places as well.”
The Biggest Risks with AI
Nick: “It’s going to be a question of not only what the tools do but exactly how people are using the tools and upleveling the ethics and the training on how people apply the technology correctly that I see as a bigger risk. It’s a leadership opportunity for people, but you don’t want to be so eager as to put this tool in the hands of people you don’t necessarily trust. That’s why being very careful about the use cases and the parameters of the use cases is going to be really important.”
As procurement teams forge ahead with AI adoption, it is clear that AI will – and likely already is – fundamentally reshape the procurement tech stack. While the potential for value creation is game-changing, we should take a measured, considered approach to risk consideration. As Lance and Nick pointed out, by establishing clear use cases, risk frameworks, and a robust governance plan, procurement can feel confident that they are not sacrificing security and safety to this revolutionary technology.
