In our latest series on how AI is being developed and applied across the source-to-pay (S2P) cycle, our analysts look at where AI is present, how it is being implemented and governed, how it is being embedded into workflows and how solution providers are answering the AI challenge. As part of this research we will include points of view from practitioner-led discussions to bring real-world perspectives to our content.
One burning question we had for practitioners was:
Across strategy development, supplier relationships and risk management, if you could have AI handle one category of decisions completely autonomously, what would that be?
We put that question to Bhavuk Chawla, who is Associate Procurement Director for wood-based packaging for the North American market at Unilever. Bhavuk is an experienced procurement, supply chain management and contract negotiation professional, specializing in manufacturing and packaging and continuous improvement in the FMCG industry.
His answer to this question was ‘risk management,’ and he explains why:
“Across the global procurement landscape, AI is reshaping how organizations plan, partner and protect their supply chains. Yet as leaders imagine a future of fully autonomous systems, an essential question emerges: Which part of procurement should AI Help with first?
“The instinctive answers of strategy development and supplier relationship management are understandable, even appealing. But both overlook a domain where AI’s strengths are not only most relevant but most urgently needed: risk management.
“In a world defined by supply disruption, geopolitical uncertainty, inflationary pressure and climate volatility, procurement risk has reached a point where manual monitoring is no longer feasible. And unlike strategic planning or relationship management, which rely on nuance, cultural fluency, creativity and influence, risk management is inherently data-driven. It is structured, time-sensitive and increasingly too complex for human-only decision cycles.
“This is exactly where autonomous AI can deliver transformative value.”
Outlining the case for AI-led risk management
Risk is a data and signal-processing challenge — AI’s natural domain
“Modern supply chains generate massive volumes of risk signals: commodity price swings, supplier financial instability, regulatory changes, port congestion, extreme weather patterns and even social media sentiment. Humans cannot realistically absorb and interpret this in real time.
“AI, however, thrives in environments built on high-frequency, high-variety data. It can:
- Detect patterns humans cannot see
- Continuously monitor global risk signals
- Update predictions instantly
- Activate mitigation plans within seconds
“AI does not wait for a meeting to escalate an issue. Risk management is, at its core, a speed game — and AI wins that game every time.”
Risk decisions can be codified far more easily than strategic or relationship decisions
“Strategies are shaped by vision, ambition and trade-offs across functions. Supplier relationships depend on empathy, trust, negotiation and interpersonal nuance. These are human spaces. Risk, on the other hand, lends itself to rules and thresholds:
- If supplier probability of default > X → trigger mitigation
- If geopolitical tension index hits Y → activate alternative routes
- If lead-time variability exceeds Z → adjust inventory buffers
“This is where AI autonomy within guardrails becomes practical. Codifiable decisions can be automated without eroding the humanity that makes procurement a relationship-driven function.”
Risk management is already algorithmic in adjacent fields — procurement is the next frontier
“In cybersecurity, finance and network management, autonomous AI systems already detect anomalies and initiate containment measures. Procurement remains one of the few lagging domains where this level of automation is not yet standard.
“But the building blocks are there. The appetite is there. And the business case is undeniable.”
AI in the wild: Real-world supply chain risk interventions
Predicting global disruptions 60–90 days in advance
“AI risk intelligence platforms have begun providing two- to three‑month early warnings for supply chain disruptions by analyzing millions of data points — from satellite imagery to payment trends. Organizations implementing these systems report 30%–40% faster response times and 20%–50% better forecast accuracy during volatility.”
Detecting supplier bankruptcy before it happens
“A global electronics manufacturer implemented an AI-powered supplier risk platform that monitors financials, news feeds and social media. It identified suppliers with high disruption probability, leading to:
- 30% fewer supplier‑related disruptions
- Improved supplier selection
- More time for procurement teams to focus on strategic priorities
“In other words, AI caught supplier failures before they cascaded into production losses.”
Predicting port congestion months ahead
“AI tools are now capable of predicting port congestion up to three months early by analyzing weather systems, vessel movements, historical port throughput and global news data. These insights help companies reroute shipments and avoid multi‑million‑dollar delays.
“This early detection capability was unimaginable five years ago.”
Real-time rerouting during weather disruptions
“Modern AI systems can rebook shipments automatically in response to weather disruptions or port congestion, acting within constraints of cost and service levels. AI agents monitor logistics flows, detect anomalies and execute corrective actions on the fly — removing bottlenecks before they escalate.
“This represents a shift from dashboard watching to autonomous exception resolution.”
Continuous monitoring of geopolitical and economic risk
“AI now ingests data on sanctions, tariffs, political instability and currency swings to maintain a real-time geopolitical risk map. Geopolitical risk categories — including trade wars, sanctions and political instability — are continuously analyzed to flag threats earlier than human analyses can.
“This supports procurement teams navigating volatile global markets.
Supplier stress detection via external data integration
“AI systems now integrate supplier financial data, capacity signals and compliance alerts to generate early warnings — often weeks or months ahead of traditional assessments.
“This enables proactive onboarding of alternatives before disruptions hit.”
Climate risk forecasting and extreme weather prediction
“AI analyzes satellite data, climate models and sensor feeds to predict how floods, hurricanes, wildfires and droughts will affect production lines or shipping routes.
“This empowers teams to move production or inventory before climate events strike.”
So what is holding procurement back?
“Despite the strong fit between AI and risk management, true autonomy remains just out of reach. The barriers, however, are solvable.
Fragmented data and limited supply chain visibility
“Risk management is only as good as the data feeding it. Many organizations still struggle with:
- Siloed ERP and supplier systems
- Incomplete supplier mapping (especially beyond tier 1)
- Inconsistent data formats
- Limited access to real-time external intelligence
“AI cannot act confidently on half an answer.”
Governance concerns and organizational readiness
“Autonomous decision-making introduces important questions:
- Who is accountable when AI triggers a disruption?
- What controls prevent cascading errors?
- How do we ensure decisions are explainable?
“Most organizations are still building the governance frameworks needed to trust AI with high-stakes operational decisions.”
AI’s difficulty in interpreting geopolitical and social context
“Risk is not always numerical. Political events, social movements and cultural dynamics may require interpretation that AI alone cannot reliably provide — yet. Human judgment remains crucial in distinguishing temporary noise from meaningful signals.”
Strategic opportunity ahead
“Giving AI full autonomy over procurement risk management isn’t about replacing humans; it’s about elevating them. When AI handles detection, escalation and pre-approved action sets, procurement teams gain the freedom to focus on what truly moves the business:
- Shaping resilient, future-proof strategies
- Strengthening supplier partnerships
- Driving innovation
- Accelerating sustainability goals
“AI becomes the always-on guardian of supply continuity partner, not a replacement.”
A future worth building toward
“The future of procurement will not be defined by who automates first, but by who automates wisely. I am of the firm opinion that risk management is the clearest, safest and most value-generating starting point for autonomous AI in procurement. It aligns with AI’s strongest capabilities, reduces the burden on human teams and offers meaningful protection against a world of increasingly unpredictable shocks.
“By embracing AI-driven risk autonomy, organizations can transform procurement from a reactive function into a proactive, predictive and resilient engine for competitive advantage.
“The question is no longer whether AI will lead risk management …
it is how soon we are ready to trust it?”
Many thanks to Bhavuk for contributing to our series. You can find more analysis, opinion and thought leadership on our dedicated AI in Procurement page.
