What Might a Future with Agentic Procurement Look Like? 

Let’s explore what the future of agentic procurement looks like with autonomous sourcing cycles, human-AI collaboration, and the evolving role of procurement professionals.

Introduction: What can we expect from agentic AI in procurement? 

It’s time to get out our crystal ball and look into the future. Not the distant future. Things are changing too fast for that. But the trends are clear so let’s focus on how agentic AI will shape the future of procurement processes over the next 3–5 years and what it means for procurement professionals. Artificial intelligence is already playing a huge role in many business functions, including procurement. However, until now, the focus has been on predictive and generative artificial intelligence. The first example of predictive AI was the JAGGAER On Time Delivery predictor, which informs companies as to the likelihood that suppliers will make deliveries on time. This is a critically important capability in manufacturing. Then came the deployment of generative AI to increase efficiency in areas such as drafting RFx and contract management. 

With the emergence of agentic AI and the shift towards autonomy and collaboration between humans and AI we are entering a whole new ball game. Right now, it would be more truthful to say that agentic AI applications are semi-autonomous. They still require a significant amount of human input, including prompts to initiate activity. But that is about to change. Let’s explore how it will change and the wider implications. 

The future of autonomous sourcing cycles: what will 2025 look like? 

Agentic AI is focused on autonomous decision-making and action. It can set goals, plan, and execute tasks with minimal human intervention. This emerging technology has the potential to revolutionize various industries by automating complex processes and optimizing workflows. 

Let’s first consider the case of a fully autonomous sourcing cycle powered by agentic AI: one in which intelligent agents, empowered by agentic AI, initiate, execute, and optimize procurement activities from end to end, without human prompts. This is already a realistic possibility. In fact, trials and testing are underway, so we could see the first real-world deployments in 2025.  

An autonomous sourcing cycle would proceed through eight stages with the various agents being controlled by an orchestration agent, which is a central AI system that coordinates, sequences, and optimizes the work of multiple specialized agents across the sourcing lifecycle. Here are the eight stages: 

1. Needs determination 

The cycle begins without human intervention. AI agents detect sourcing needs through real-time integration with ERP or inventory systems, demand forecasting algorithms identifying upcoming shortages, and contract lifecycle monitoring to identify expiring agreements or performance risks. The cycle is triggered by an AI agent, typically because it notices that a specific component’s stock is below a safety threshold; but there could be some other trigger such as the agent noticing that contract pricing is out of line with market rates. 

2. Sourcing strategy design 

An AI agent selects or configures a sourcing strategy appropriate to the context (spot buy, reverse auction, multi-tier sourcing, etc.) It then formulates risk-weighted decisions based on supplier concentration, ESG ratings, tariff exposure, etc., with consideration of total value, not just unit cost (e.g., resilience, carbon impact). 

3. RFx creation and issuance 

Next, the system auto-generates RFQs, RFPs, or e-auction configurations tailored to the category and spend. An agent autonomously selects suppliers using pre-vetted pools enriched with external market intelligence and ESG/compliance scoring. The system then issues RFx packages and manages Q&A interactions autonomously. 

4. Supplier evaluation and shortlisting 

The agentic AI scores responses against weighted criteria and simulates different award scenarios (based on cost, risk, and ESG etc.) It flags anomalies or compliance red flags using past performance and predictive analytics. If scores are close or a human override threshold is triggered, it may escalate to a user. Otherwise, it moves forward. 

5. Autonomous negotiation 

Using pre-defined constraints and goals, the negotiation agent engages in digital negotiation with suppliers (if the supplier platform supports machine-to-machine protocols) and optimizes for total value using price, delivery, flexibility, and terms. (It is likely that for the foreseeable future this will only be the case for non-strategic and non-critical categories, with human intervention in other cases.) 

6. Contract drafting and execution 

Selected suppliers are auto-onboarded via a digital workflow and offered a contract with clauses adapted from legal playbooks and specific to the negotiated terms. The contract could be signed via e-signature or else flagged for legal review. 

7. eProcurement execution and monitoring 

The PO is issued and delivery schedules tracked. Payment is triggered and early payment discounts are optimized automatically. Supplier performance is continuously monitored against SLAs, ESG KPIs, and risk indicators. 

8. Continuous post-award monitoring and re-sourcing 

If the supplier underperforms or a disruption is detected (e.g., following a geopolitical event, or cyber breach), the AI agent automatically initiates a re-sourcing process. Lessons are fed into the training set for future optimization. 

Whereas much of the above can be, and already is, automated, with agentic AI the cycle moves procurement from tactical automation to autonomous orchestration, delivering agility, cost savings, and resilience at scale. The orchestration agent is rather like an air traffic controller for intelligent sourcing agents. Just as an ATC officer doesn’t fly the planes, the orchestration agent does not perform actions such as issuing RFQs and negotiating, but rather ensures that these actions are scheduled, monitored, rerouted if needed, and executed safely from source to pay. 

Human-AI collaboration in procurement: the new work model 

This description of an autonomous sourcing cycle may leave procurement professionals asking what role (if any) they will have in the future. The fact is that some jobs will go, because autonomous procurement will achieve far more, in far less time, than humans can — but only in certain areas. AI agents will take care, for the most part, of the mind-numbingly boring manual tasks such as filling in forms, monitoring activity and writing reports. It is unlikely that these activities will be completely taken over by agentic AI any time soon, however. Moreover, human intervention will continue to be necessary in many cases and essential in others, such as the negotiation of awards in sensitive categories. 

In the short term there will be skill mismatches. Employees displaced by automation may not be immediately ready to take on more creative or strategic tasks. The World Economic Forum, McKinsey, and OECD all note that while AI could augment human roles, it also demands upskilling and organizational change that many firms aren’t ready for. Moreover, the shift towards more strategic work may have downsides, unless people are adequately trained. Strategic roles often come with increased responsibility, and therefore more stress, as there is pressure to deliver insights quickly (especially in data-rich environments). 

The lesson is that companies need to approach this issue with care and with close attention from HR and corporate leadership. Technology alone will not drive a strategic reallocation of human resources. You need changes in KPIs, incentives, and workflows. Without proper guidance, middle management may simply reallocate freed-up capacity toward equally tiresome activities (for example, fewer workers managing a greater volume of sourcing events). However, with strong leadership and structured initiatives (such as supplier innovation days, risk war rooms, ESG scorecards) procurement can be made a more interesting and rewarding place to work in future. 

The role of procurement professionals in an agentic AI world 

So, there must be an evolution of the procurement professional’s role, a shift from tactical roles to more strategic and advisory positions, which will benefit procurement professionals. But the shift requires intentional design, reskilling, and cultural change, which won’t happen by default. Procurement professionals will manage and oversee AI-driven decisions while ensuring ethical considerations, compliance, and human oversight, but they must be equipped to do this.  

An emerging role in this environment is an “AI enablement specialist for procurement.” These individuals or teams will act as translators, trainers, and guardians, ensuring that AI adoption in procurement is not only technically sound but also ethically, legally, and operationally robust. Key competences they will bring include: 

• Change management: Helping teams shift from manual processes to AI-augmented workflows 

• Ethical oversight: Ensuring decisions made by AI (e.g., supplier scoring) don’t reinforce bias or undermine DEI goals 

• Compliance alignment: Making sure AI-driven decisions adhere to regulatory and internal audit standards 

• Human-in-the-loop design: Clarifying when and how humans should intervene or override autonomous processes 

• Skill development & coaching: Supporting procurement professionals in understanding AI outputs, interpreting risk signals, and acting on recommendations 

Without such support, procurement staff may not trust or understand AI-driven insights and are likely to resent the shift in expectations. Moreover, there is a risk of “black box” procurement emerging, where decisions can’t be explained and it is unclear who is accountable for decisions. Without human oversight, legal exposure increases in some situations, for example, if AI filters out suppliers in a way that violates fair competition laws. 

It’s already starting to happen in some industries. Pharmaceutical firms are bringing in ethics and governance officers specifically to review AI-powered supplier risk algorithms for bias or false negatives. 

Key benefits of agentic procurement and the path to implementation 

Agentic AI can deliver cost savings, efficiency, and accuracy in procurement. Therefore, it is inevitable that it will be implemented in many sectors, first in large firms but eventually becoming ubiquitous. If implemented with care, it will reduce errors and biases in vendor selection and contract negotiation, reduce maverick spend, eliminate manual effort and increase throughput, among other benefits. 

To ensure a smooth path to implementation, organizations should be developing new certifications and training programs in AI-augmented procurement ethics and governance. They should create the appropriate new roles and job descriptions such as AI Workflow Architect, Procurement-AI Liaison, and Cognitive Process Auditor. And they should engage third-party professional services specialists, for example from management consultant firms and from technology providers such as JAGGAER, who will offer tailored guidance on safe, strategic AI integration. 

Key takeaways: preparing for the future of agentic procurement 

Agentic procurement is coming, and it will bring many benefits to companies that embrace it successfully. We are getting closer to the concept of autonomous commerce, whereby buyers and suppliers can entrust a lot of the routine work to software agents. But there will always be a need for the human in the loop and organizations need to prepare procurement professionals for the transition. 

With the advent of agentic AI, procurement teams will need to embed new roles to ensure AI systems are fair, transparent, accountable, and aligned with organizational values.