Title

Generative AI and Procurement: Opportunities, risks, and approaches

Section
Insight
Summary

The past year has seen Generative AI, or perhaps rather the idea of it, being brandished as a revolutionary force across all business areas, with procurement and supply chain no exception. Here, we’re taking a step back and looking at the practical: what will the future of artificial intelligence and procurement look like in reality? 

This article examines how we expect AI advancements to impact tasks, outputs, and procurement professionals’ roles; it also outlines the risks and recommended approaches to adopting AI tools. No doubt AI can improve the productivity and strategic value of procurement – but only for the organizations and procurement leaders that follow the right processes to harness its potential and mitigate the associated risks.

Artificial Intelligence in Procurement

Automation will be a key driver for the adoption of artificial intelligence, particularly Large Language Models (LLMs), in procurement. A growing number of AI tools and integrations are streamlining routine tasks that have typically consumed a large portion of a team’s time, including:

  • Document generation: AI can draft routine documentation such as purchase orders and contracts with reduced likelihood of errors, offering productivity gains across the procurement lifecycle.
  • Data entry and analysis: AI systems excel in data processing. They can process extensive datasets to extract relevant insight for decision-making, which is particularly useful in analyzing supplier performance and risk factors.
  • Initial vendor negotiations: AI can manage routine negotiations based on predefined criteria, allowing for the automation of initial discussions and negotiations of basic parameters. This opens procurement professionals up to focus on more strategic and nuanced aspects of their roles, such as complex negotiations and relationship management.
  • Market trend monitoring: AI algorithms can monitor and analyse market trends in real time, arming procurement professionals with up-to-the-minute insights to support proactive and ongoing decision-making based on the latest market developments.

Long-term, the use of AI in procurement will extend beyond simple automation. We expect advanced use cases for AI in procurement to include:

  • Dynamic negotiations: Armed with advanced natural language processing and negotiation algorithms, AI will support data-driven and intricate negotiations, including managing complex contractual terms and intricate pricing structures and tailoring agreements to meet the specific needs of both parties. Understanding and adapting to the subtle nuances of negotiations will mark a significant evolution in AI's role within the procurement domain, with a combination of artificial and human intelligence driving better results.
  • Strategic procurement through predictive analytics: Integrating AI-driven predictive analytics into procurement will create a strategically dynamic function; they will predict market trends and foresee potential supply chain disruptions. These systems will leverage comprehensive data analysis to recommend proactive adjustments to procurement strategies, helping businesses to stay ahead of the curve in rapidly changing environments.
  • Holistic sustainability assessments: AI tools will play a pivotal role in holistic sustainability assessments. Beyond evaluating the environmental impact of suppliers, these tools will analyze diverse datasets related to ethical business practices, social responsibility, and overall corporate values. Organizations will be able to make data-driven decisions to align their procurement processes with their broader corporate social responsibility and environmental sustainability goals.

The automation of routine tasks will increasingly liberate procurement professionals from administrative and repetitive tasks – a shift that will enable a greater focus on higher-value tasks, such as strategic planning, relationship building, and creative problem-solving.

A common concern surrounding AI is its impact on jobs. However, the human elements of procurement, such as building and maintaining relationships with suppliers and cross-functionally within the business, will remain crucial. In the long term, as AI handles an increased number of tasks, procurement roles will demand more human-centric skillsets such as nuanced decision-making, emotional intelligence, and relationship building and management. In other words, we expect a shift in the skills needed for procurement functions to achieve success. Automation may make some skills less important while boosting the demand for tech proficiency, strategic thinking, and collaborative abilities. Adapting will be key.

Ultimately, technology and data can only take you so far. Procurement should focus on using AI as an ally to facilitate better supplier outcomes and navigate complex negotiations. It is a combination of both human and artificial intelligence that will drive best outcomes – and, as with any transformation, the right procedures and training are needed to make this happen.

 

Implementing AI: risks, challenges, and recommended approaches 

The introduction of AI into businesses presents opportunities, but it also poses risks that organizations must tackle upfront. Much of the risk minimisation should centre around laying the groundwork for appropriate usage ahead of implementation.

Don’t underestimate the importance of training and education

It is very likely that there will be varying levels of understanding and experience of AI tools across the organization. The risks posed by the misuse or ill-informed use of AI (as well as the potential opportunities created by appropriate usage) demand the implementation of formal training structures to build a common understanding of the implications of AI usage.

Take AI chatbots, such as ChatGPT or Google's Gemini. Employees must be warned against inputting sensitive or confidential information (if not a private instance) – the business world has already seen instances of information leaks via such tools. Before making any AI tools available to your teams, make sure that they are educated on these risks and how to use them safely. In addition, your teams should understand that AI outputs are not flawless; human judgement should always be applied to validate the results, particularly if the results are being used for a meaningful business decision.

Data and governance: build a strong foundation

The output of AI tools will only be as good as the available data. Incomplete or low-quality data creates performance and ethical risks, such as biased supplier selections, misjudgement of market trends, or incorrect assumptions about supplier performance. Put checks in place to ensure the data being used is accurate and ready to be used, or risk making business decisions based on misleading information. This also raises the “build” vs “buy” question, given third-party tools can be a black box; we suggest organizations weigh up the importance of a tool’s role and output versus the available transparency and the required investment.

Put the appropriate governance in place ahead of tool implementation or integration, with a strong understanding of use cases, risks, protective measures, and data and content access. In addition, make sure you understand your cost models and resource needs before you get started. Involve senior business leaders and the appropriate stakeholders (such as legal, tech, finance) in the decision-making and implementation processes. Evaluate whether you have the right capability in-house or need support from a third-party expert.

Start small with a phased roll-out in any instances of AI use, and make sure to continuously review usage and impacts.

Getting started

We stand at the cusp of a procurement shakeup, where Generative AI graduates from the realm of the buzzword and begins to become a strategic ally for procurement. If embraced and correctly deployed, artificial intelligence could boost procurement’s journey to further becoming a strategic partner to the business. But procurement teams that are slow to adapt risk falling behind and losing out. Implementation and adoption, as well as attitude and skillset evolutions, can’t happen overnight – the key is to start now, towards an era of buying better.

Download PDF