AI System Procurement Best Practices

#AIProcurement Is Redefining How Organizations Buy, Build, and Manage Technology Here are the insights from my latest guest spot with Vishal Patel on the Ivalua #LoveProcurement Podcast: ----------------- KEY TRENDS & INSIGHTS 1️⃣ You’re Not Buying a Product — You’re Buying a Behavior - Acceptance criteria must include edge cases, hallucination risk, bias risk, and failure modes. - You must evaluate model behavior, not just technical specs. - Vendor evaluations must incorporate “black-box testing” and scenario trials, not just security questionnaires. 2️⃣ The Most Important Contract Clause Isn’t Price — It’s #ModelChanges Smart procurement teams negotiate: - Model change notifications - Update sandboxes - Version pinning - Right to revalidate - SLAs for output quality, not just uptime 3️⃣ Procuring AI Requires Cross-Functional Governance — Not Just a Contract AI touches everything, and all parties must participate in governance upkeep: - Customer data - Internal knowledge systems - Decision-making - Intellectual property - Regulatory exposure 4️⃣ Your Existing Evaluation Framework Is Probably Outdated AI evaluations require new categories of due diligence: - Training data provenance and licensing - Model lineage - Fine-tuning risks - Bias, safety, and security testing 5️⃣ AI Procurement Is Moving from a Cost Focus to a Capability Focus The real differentiators are: - Data inputs and outputs - Integration depth - Customization options - IP rights - Safety and security guarantees - Performance -------------------- THE BOTTOM LINE Leaders who update their procurement frameworks now will innovate faster, govern more effectively, reduce long-term risk, and curb losses. Those who don’t? They’ll end up with shadow AI, compliance problems, and expensive rework down the road. ------------ Watch the full episode here:  https://lnkd.in/emsvDjZK AI Procurement Lab

Want AI in Procurement? Don’t Skip the Basics! (A lesson from CPO’s "aha" moment) CPO: How fast can we deploy AI to cut costs & predict risks? Me: Let’s talk about your foundations first. Think of "deploying AI" like building a house: 🚫No foundation (data)-Your house collapses. 🚫No walls (processes)-Rain floods in. 🚫No electricity (skills)-You’re stuck in the dark. 🚫No plumbing (analytics)- Things get messy. His reality: 📉 Data was scattered across 7 systems 🧑💻 Team had zero analytics training. 🔄 Processes were manual, inconsistent & slow. Sound familiar? You’re not alone. Procurement Excellence | MAR 2026 - Deploying AI in procurement without clean data or a trained team is like building a skyscraper on a quicksand. The 5-Step Pyramid for AI-Ready Procurement (Start at the bottom!) #1. Clean Data ↳Get your basic facts straight. ↳No typos in supplier names, no duplicate orders, all spend tracked in one place. #2. Smooth Processes ↳Make your workflows simple and consistent. ↳Everyone follows same steps to approve purchases or sign contracts. #3. Trained The Team ↳Training on data analysis via use of new tools ↳AI helps humans it doesn’t replace them. Scared or confused teams won’t use it. #4. Basic Analytics ↳Use data to spot trends and measure success. ↳You need to walk before you run. Master simple insights before predicting the future. #5. AI ↳Let tech do complex tasks automatically. ↳Predicting shortages, negotiating prices, or finding risks before they happen. AI is the peak of procurement evolution—but you can’t jump straight to the summit. The Hard Truth AI isn’t a quick fix. It’s the climax of a journey Your Action Plan: ✅️Audit your data. ✅️Map processes & fix bottlenecks. ✅️Train your team on data literacy. ✅️Start small using basic analytics. ✅️Then and only then pilot AI. Skip steps & AI becomes expensive hype. Build step-by-step & it changes everything. What are other considerations for deploying AI in procurement? ♻️ Repost to help someone in your network 🔔 Follow Frederick for more hard truths about AI in business. #Procurement #AI #DigitalTransformation #DataDriven #Leadership #Maslow #Innovation

📜 Every time a company acquires an AI system, it must ensure legal due diligence and a well-structured contract, especially for high-risk use cases. To support this complex process, the European Commission has recently updated the EU Model Contractual Clauses (MCCs) for the Procurement of AI Systems. Although originally drafted for public entities, private organizations can also adopt or adapt the clauses when acquiring or developing AI systems. They serve as a valuable benchmark for any company, especially as the EU AI Act, despite its detailed scope, still leaves room for interpretation regarding specific contractual requirements. The revised MCC-AI are designed to align with the new AI Act and are available in two formats: 1. Full Version (High-Risk): Tailored for AI systems classified as high-risk under the AI Act, such as those used in recruitment, credit scoring, education, or healthcare. 2. Light Version (Low/Moderate Risk): A simplified alternative for AI systems that do not meet the high-risk threshold but may still affect fundamental rights or safety. ⚖️ Key Legal Provisions – Full Version (High-Risk AI Systems): 1. Technical Requirements: Obligations related to the system’s accuracy, robustness, and cybersecurity. 2. Supplier Responsibilities: Requires implementation of quality management systems and conformity assessments. 3. Data Governance: Clearly defines rights and obligations over the datasets used to train and operate the AI system. 4. Audit & Accountability: Grants public buyers the right to audit the supplier to verify compliance. 5. Indemnity Clauses: Suppliers must indemnify the buyer for any violations of intellectual property or data protection rights. ⚖️ Key Legal Provisions – Light Version (''Low/Moderate'' Risk AI Systems): 1. Transparency & Documentation: Suppliers must provide clear documentation about the system’s design, functionality, and purpose. 2. Data Governance: Sets out standards for data use and protection within the context of the AI system. 3. Exemptions: Unlike the high-risk version, it does not require formal conformity assessments or a full quality management system—reflecting a lighter regulatory burden. 🚨 Non-Binding Nature: The MCC-AI are non-binding templates designed to be tailored, adapted and annexed to broader procurement contracts. 🚨 Scope: These clauses focus specifically on AI compliance and the AI Act, without addressing unrelated contractual areas such as Data Protection, IP ownership, SLAs, or payment terms. Link for the updated Model Clauses: https://lnkd.in/eHzJtis7

The power of the NPI community - wow! We held an AI special session with our advisory board last week - and we surfaced some super-effective AI contractual approaches. This board is comprised of the smartest and most sophisticated procurement leaders on the planet - and I am not exaggerating. They are just amazing. Think 40+ CPOs and Heads of IT Sourcing coming together and exchanging ideas. We protect this group’s identities due to our fierce privacy culture. But just know, these execs are the real deal. We conduct these sessions for the benefit of our customers - but I’ll share some nuggets here: 1. Consumption Transparency: Make It a Contractual Obligation Demand explicit documentation of which tasks consume credits or tokens, at what rate, and how rates vary by model tier or environment (sandbox vs. production often carry different multipliers). 2. Overage Protection and Hard Spending Caps Overage exposure is where consumption models get expensive fast. Negotiate pre-agreed overage pricing locked at or below your contracted per-unit rate, hard caps or throttles that prevent runaway consumption. 3. Price Lock on Per-Unit Consumption Rates If you're committing to multi-year consumption deals, lock per-credit or per-token pricing for the full term, not just Year 1. Add Most Favored Customer language so that any better per-unit rate made available to other buyers must be extended to you. 4. Anti-Repricing Protection for Currently Free Features This is an emerging and documented vendor tactic: launch AI capabilities at zero cost to drive adoption, then price them once workflows are built around them. Strike any language that gives providers the ability to change pricing mid-term, or narrow it to a specific, measurable trigger. 5. Pilot Periods Before Volume Commitments Consumption is nearly impossible to forecast before you have real usage data. The most defensible approach: negotiate a 9–12 month pilot with contractual decision rights at the end: proceed, reduce, or exit. 6. Shorter Terms with Structured Opt-Outs Multi-year AI commitments are increasingly difficult to justify when the technology and vendor landscape can shift materially within 12–18 months.   7. Cross-Document Contract Integrity Every new order form, addendum, or module addition should include an explicit subordination clause confirming it does not override data-use or confidentiality provisions elsewhere in the agreement. Every one of these provisions is easier to obtain before you're operationally dependent on the platform. Once teams have built workflows on consumption-based features, walk-away leverage disappears. Vendors know it. The contract is the only protection you have and the time to negotiate it is now.

Procurement is where AI governance gets real. And the EU just delivered one of the most practical tools yet to help public authorities navigate it responsibly. 📘 The MCC-AI (Model Contractual Clauses for the Public Procurement of AI) is a guidance framework developed by the European Commission’s Public Buyers Community, designed to help public institutions embed the AI Act’s risk-based requirements directly into procurement contracts. What’s remarkable is not just the content — it’s the intent: ➡️ To operationalize compliance, ➡️ Reduce vendor lock-in, ➡️ And build capacity through learning-by-doing. 🧩 What’s inside? There are two versions: MCC-AI High-Risk – for AI systems falling under Chapter III of the AI Act MCC-AI Light – for non-high-risk systems, with adaptable clauses Covered areas include: ✔️ Risk management and conformity assessment ✔️ Data governance and access to datasets ✔️ Logging, documentation, and human oversight ✔️ Transparency obligations and individual-level explainability ✔️ Rights over public and supplier AI datasets ✔️ AI registry participation and post-contract handover The framework even provides annexes for defining intended purpose, technical specs, and robustness thresholds — creating a structured, modular way to implement AI-specific contract language . 🙌 Full credit to Jeroen Naves (author) and Anita Poort & Ivo Locatelli (reviewers) for producing a working document that strikes a rare balance: it’s grounded in the AI Act but driven by real implementation needs. 💡 Why it matters? Because policy impact starts with contracts — and this tool shows what it looks like when a regulator commits to clarity, reuse, and learning as core design principles. #AIGovernance #AIProcurement #ResponsibleAI #AICompliance #AIAct === Did you like this post? Connect or Follow 🎯 Jakub Szarmach Want to see all my posts? Ring that 🔔.

AI in Global Supply Chains — Part 3: Sourcing & Procurement Last week, Tesla gave Syrah Resources more time to meet graphite anode specs under a multi-year supply deal—proof that capability, qualification, and terms can make or break a sourcing bet. In July, Apple committed $500M to a U.S. rare-earth magnet partnership with MP Materials—an example of value-led sourcing that blends cost, resilience, and responsible materials. This is how I have been able to work with clients on similar challenges, using AI. 1️⃣ Market intelligence & should-cost (research) Map the landscape in hours: capabilities, certifications, footprint, and rough capacity signals. Pull public price lists, tariff/FX, and input bills (materials, labor, energy) to build should-costs and a risk-adjusted total landed cost. Screen for responsible sourcing and safety practices without slowing the process. 2️⃣ Shortlist, outreach, and RFI/RFP Cluster and de-duplicate suppliers; auto-draft multilingual RFIs; normalize replies (units, currencies, terms). Score proposals across capability, capacity, quality, cost, lead time, logistics, and risk—not just price. Run scenarios (MOQ changes, dual-source, regional mix) before you invite to RFP. 3️⃣ Negotiate for value (not price alone) Use multi-objective trade-offs: tooling amortization, yield guarantees, service levels, buffer stock/VMI, payment terms, FX/pass-through rules. AI copilots surface give-gets and simulate outcomes (service, cash, contribution margin) so you walk in with a plan, not a number. Use structured events (ranked bids or multi-attribute auctions) when appropriate. 4️⃣ Contract draft, terms negotiation, and redlining Clause libraries + AI redlines flag deviations, propose fallbacks, and summarize changes by risk. Link SLAs, quality plans, and service credits to measurable data; push into CLM so obligations don’t get lost after signature. 5️⃣ Onboarding, pilot, and ramp Digitize onboarding (tax, banking, compliance), connect EDI/API, and run first-article/PPAP or equivalent. Stand up a 30-60-90 day ramp plan with early-warning KPIs. ➡️ Bottom line: AI turns sourcing from a price hunt into a repeatable, value-optimized system. Next in the series: AI for Design for Manufacturing and Circularity.

This isn't about surviving layoffs. This is about positioning yourself as the one who makes technology investments actually work. IT procurement pros, you're holding the keys to what comes next. Every AI tool they're betting on... Every automation "efficiency gain" they're promising shareholders... You know they can't handle that without you. Time to make some moves... 1. Your stakeholders just lost half their team. They're drowning. Don't wait for them to come to you... Show up with solutions. "Let me help you figure out how to maintain capability within budget." 2. Build your AI procurement playbook before someone else does. Most companies don't know how to evaluate AI vendors. Create an AI evaluation framework. Make yourself the go-to expert on AI procurement in your organization. Because I promise you – if you don't, some consultant will, and they'll charge 10x what you would've saved them. 3. Quantify everything you touch. In a "do more with less" environment, the people who can prove their value in dollars don't get cut. They survive the crunch. Document every vendor consolidation, every cost avoidance, every contract you kept from auto-renewing at list price. Build your case now, before someone starts asking questions. Know this: your peers are your early warning system. Use that. Share what's working. Learn what's not. Build your intel together. This is the wild west. Someone needs to write the playbook. Why not you? Position yourself as the strategic partner who ensures those AI investments deliver. The companies that are cutting jobs are the same ones increasing technology budgets. They need you to make sure those dollars deliver the efficiency they promised—and to make yourself indispensable in doing so. What's the one thing you're doing this month to prove your strategic value? If it's getting harder, you are getting closer. Keep pushing, Procurement.

💡 4 Best Practices for Forwarders Rolling Out AI 🤖 Adopting AI can feel intimidating at first, but it doesn’t have to be. From what I’ve seen across implementations, if you ensure the AI solution fits your workflows, start simple, and use the implementation as a forcing function to clean up existing processes, you’re setting yourself up for success. Here are four best practices I’ve seen to maximize success 👇 1️⃣ Trial before signing. AI automation can look great in demos, but the real test is whether it performs with your documents, systems, and processes. A short trial is the best way to find out before making any commitments. 2️⃣ Check for hidden manual work. If your ops team has to label data or train the model, the work hasn’t gone away, it’s just been renamed. 3️⃣ Start simple. You don’t need to automate everything on day one. Start with one workflow in one office, learn fast, then expand. Don't let perfection prevent starting. 4️⃣ Fix broken processes first. AI can’t turn a broken workflow into an efficient one. The best teams use implementation as an opportunity to clean up how work gets done today, then layer automation on top.