.png)
We brought together seven procurement and technology professionals to provide their honest opinions on AI - covering everything from cautious realism to hands‑on optimism. Their perspectives converged on one central truth: AI has real value - but only when applied with clarity, context, and care.
1. Ground AI use cases in real procurement pain points
Don’t start with technology. Start with the problem.
One professional explained, "Prompting is just structured input. Treat it like a form."
Rather than chasing cutting-edge capabilities, many procurement teams can unlock impact by solving persistent challenges like:
- Cleaning and classifying messy supplier data
- Automating invoice matching and approvals
- Simplifying supplier onboarding or RFx processes
Actionable steps:
- Conduct a workflow audit to identify bottlenecks or repetitive tasks
- Run pilot projects using AI to handle low-risk, repetitive work
- Set measurable outcomes (time saved, error rates reduced, etc.)
2. Treat AI like outsourcing - build governance, not blind trust
AI is not a junior employee - it's an unpredictable partner that can make confident mistakes.
One contributor warned, "Think of agentic AI as outsourcing a task - you wouldn't do that lightly."
Procurement leaders must treat AI agents like third-party vendors, with access controls, risk mitigation, and oversight built in.
Actionable steps:
- Implement permissioning and audit logs for AI-driven tasks
- Establish review protocols for AI-generated outputs
- Work with IT and legal to ensure data governance policies extend to AI use
3. Build internal AI literacy before chasing sophistication
Your team doesn't need to be AI experts, but they do need to understand the basics.
A panelist noted, "The future is orchestration - tools directing tools, not one agent doing everything." Enabling teams to prompt well, validate outputs, and think critically about when (and when not) to use AI is more valuable than deploying the flashiest tool.
Actionable steps:
- Run internal training on AI basics, prompt design, and risks
- Encourage experimentation with structured guardrails
- Assign AI champions or stewards across departments
4. Beware the hype gap - match use cases to maturity
Don’t let vendor marketing set your expectations.
Another expert cautioned, "If you don’t have data discipline, you’ll get nonsense out of even the best model."
While some agentic capabilities sound impressive, they often underdeliver without strong inputs, robust process design, and clear evaluation criteria.
Actionable steps:
- Ask vendors for proof of outcomes, not just feature lists
- Ensure your team can measure model performance and identify failure modes
- Avoid using AI for high-stakes decisions without human-in-the-loop verification
5. Use AI to accelerate modernisation - not avoid it
AI won’t fix broken workflows - it will expose them.
One speaker said, "AI reveals the mess in your processes - that’s a feature, not a bug."
Rather than automating chaos, AI should be used as a catalyst to simplify, standardise and digitise legacy practices.
Actionable steps:
- Use AI adoption as a reason to clean up data and redesign workflows
- Focus on modular, reusable processes that AI tools can plug into
- Prioritise data standardisation across supplier interactions
6. Drive AI strategy from the top with clarity and patience
AI transformation is a cultural and strategic shift - not a side project.
One professional shared, "There is no shortcut to getting ready for AI. The work has to be done."
Leadership must align around what AI is (and isn't), define acceptable risk, and support the reskilling required to use it well.
Actionable steps:
- Create a shared vision of AI’s role in procurement across your leadership team
- Invest in cross-functional buy-in, not just tools
- Build in checkpoints for reflection, iteration, and learning
The bottom line
AI is no silver bullet - but it is a powerful lever. For procurement teams willing to do the groundwork, stay realistic, and centre their strategies on actual needs, AI can enable faster decisions, cleaner data, and stronger supplier relationships. Just don’t forget: the real transformation is human-led.
This article draws from the panel discussion: “Disentangling AI in ProcureTech: Singularity, Snake oil or Somewhere in-between?" featuring 7 procurement experts with different perspectives on AI - for the full download - watch the on-demand discussion here.
Smarter sourcing. Faster insights.
Leverage AI-powered eSourcing with DeepStream’s new AI analytics module - featuring conversational analytics, shortcut analysis buttons, and bespoke dashboards.
From complex award scenarios and executive summaries to supplier feedback reports and those left-field C-suite requests, you can summarise your data any way you need, in seconds.
