4 real AI shifts that will define procurement in 2026

There is endless media speculation over new AI breakthroughs and the "ifs and whens" of Artificial General Intelligence (AGI). While their bold headlines capture the imagination, they often distract from the real, tangible changes happening right now in areas like procurement, where AI already delivers high returns on investment.

In a recent webinar, Rhonda Spraker Griscti, Executive Director of Agile Sourcing at Bristol Myers Squibb, noted that thanks to AI, their RFP timeline had dropped from six-nine months to 27 days and had eliminated five months of cycle time. They are now processing ten times more RFPs than before, an outcome, she said, almost unheard of in software rollouts.

Outcomes like these fuel my prediction that the most significant shifts next year will be practical and deeply integrated into the core operational fabric of business. So ditch the sci-fi fantasies! The real story is how AI is driving a new enterprise focus on specialization, transparency, and trust in order to deliver ROI.

1. Smarter, Specialised Models will Trump Large

While large language models (LLMs) will certainly get better, we should expect incremental improvements rather than revolutionary leaps. Future models will likely hallucinate less, handle function calls more effectively, and offer larger context windows for less reliance on retrieval-augmented generation (RAG). The real progress by 2026, however, will come from improving how we integrate and apply the powerful AI capabilities that are already available.

The core of this trend is specialization. Companies will increasingly complement huge, general-purpose LLMs with smaller, specialized models. These domain-tuned models are more efficient and perform better on specific tasks because they are trained on data grounded in actual business realities, not just on broad, ungoverned internet text.

This shift makes a company's unique, proprietary data its most critical strategic asset. Building a "data moat" will become one of the most important trends for any B2B company. Software providers must own their data, understand it, and use it to train specialized models that create a distinct competitive advantage. Critically, this must be done ethically and with respect for customer privacy.

2. "AI Transparency" will Split in Two Essential Parts

Transparency will emerge as a non-negotiable battleground for trust in 2026, but with two distinct facets that businesses must embrace to stay competitive.

The first is governance transparency. Every company using AI will need to disclose how it manages these systems, including its guiding principles, data handling policies, and safety measures. Large enterprises are already demanding this from their suppliers, sending detailed AI questionnaires and excluding those with weak or incomplete answers. To compete, organizations must back up their governance policies with hard evidence: audits, documentation, and verifiable architectural practices that ensure customer safety.

The second is decision transparency. This means making it clear when AI is involved in a decision and, crucially, being able to explain why the AI reached a particular conclusion. For example, when an AI agent provides an answer, the system should indicate the source and show what data or reasoning supported the response.

3. AI Agents will gain more Independence, and Personalities

AI is moving beyond simply completing commands and is evolving into autonomous agents that can manage daily work, particularly in complex fields like procurement. In 2026, the automation of negotiation in procurement will become more widespread. Because effective negotiation relies heavily on access to accurate, historical data, platforms with deep procurement datasets are uniquely positioned to automate these interactions. And they will do so in personalized ways, adopt for example a more collaborative or more competitive tone, depending on the context.

It's important to stress that this automation isn't about replacing people; it's about solving a critical business problem. Many organizations still face large backlogs of sourcing events that never receive proper analysis or negotiation. Delegating routine work to AI helps reduce that backlog, improve spend coverage, and give teams greater strategic control.

4. Get Ready for AI's Version of a Security Badge

As AI becomes more integrated into business operations, governance and compliance standards are emerging to ensure it is used responsibly. The most important of these is ISO 42001, the new international standard for AI management systems. Its purpose is to provide clear principles that organizations must follow to ensure AI is developed and deployed safely, responsibly, and without bias.

While adoption is still in its early stages, its future importance cannot be overstated. It is poised to become a common badge of credibility for AI software, much like ISO 27001 did for information security. It signals the maturation of the AI industry and will be essential for building lasting trust.