2026 Decision Intelligence Benchmark — Special AI Report

The State of AI in Supply Chain

How supply chain leaders are adopting AI — or not — to achieve their business objectives

Table of Contents

This report is interactive. Use the controls to filter all charts by role level. Charts update instantly. Hover over bars to see exact values. Click All respondents to return to the overall view, which is the view reflected in the written narrative.

View by segment
Executive Summary

Supply Chain Ranks High in AI Ambition — Last in Proficiency

Supply chain enters 2026 ranked 2nd of 11 benchmark functions on expected daily AI use, 2nd on actual daily adoption, and 2nd on automation outlook. On three of the six AI vectors tracked in this benchmark, only one other function scores higher. Supply chain is tied directly to cost, service levels, and time to market — and the function's leaders are investing in AI accordingly. What they do not yet have is the team proficiency to match that posture. On that single vector, Supply chain ranks last of all 11 functions.

The proficiency gap is mixed and not evenly distributed. For example, directors unanimously rate their teams as beginners on enterprise AI — meanwhile senior leaders give teams much higher ratings. Members describe a function where executives are setting ambitions that the teams closest to implementation are not yet positioned to meet. Across all three AI types, "beginner" is the dominant competency rating – even for widely accessible tools in public GenAI.

The mistrust data adds context. Inaccurate or hallucinated outputs are the primary trust concerns, and data privacy is right behind it — the two highest mistrust drivers in the function, both running well above most peer functions. For supply chain, these are not abstract concerns: a flawed demand forecast or an incorrect supplier risk score carries direct operational cost. Trust in AI, like proficiency, follows use — and both are still forming in this function.

Different Perspectives on Different Types of AI

For the purposes of this report, and to better understand how leaders think about different kinds of AI, we grouped tools into three categories:

Public Generative AI
Broadly available tools such as ChatGPT, Claude, Gemini, and similar public-facing generative AI platforms.
Enterprise Platform AI
AI features embedded in vendor-supplied systems, such as ERP platforms, planning suites, or other enterprise software.
Company-Owned AI
Proprietary, internally governed, or organization-controlled AI tools built, licensed, or configured specifically for the company or its teams.
Part 1: AI Adoption

Supply Chain Sets the Highest Expectations

Supply chain leaders have set ambitious targets for daily AI use across all three categories — and the adoption gap is widest where leaders expect the most usage.

Highlights from the data
  • Enterprise AI expectations lead, actual usage trails: 67% of supply chain leaders expect daily enterprise platform AI use, with expectations for daily public GenAI use close behind. Enterprise AI has the largest expectations vs. actual usage gap at 34 percentage points.
  • Public GenAI is the most-used type — but middle of the pack across functions: 50% of supply chain leaders report daily public GenAI use — the highest among the three types of AI for the function and above the cross-functional average of 48%, but 6th out of 11 functions surveyed, suggesting its public GenAI use is middle-of-the-pack. Daily use of company-owned AI (33%) ranks 3rd out of 11.
  • Directors and senior leaders diverge sharply on company-owned AI: None of the directors expect daily use of company-owned AI, signaling fundamentally different reads on the anticipated value of this technology.

Expectations of AI Use

Supply chain leaders carry some of the highest AI use expectations in the benchmark. 67% expect their teams to use enterprise AI daily — the highest expectation across all three AI categories for this function and third out of the 11 functions surveyed.

The role-level cut on company-owned AI represents the sharpest divergence in the data. Senior leaders (VP and above) expect 67% daily use. Directors expect none. This gap does not appear for public or enterprise AI, where both groups land at nearly identical figures.

Expected Daily Use by AI Type
Q: What is the frequency of use you expect from your team for each AI type?
(N=17) · "N/A" responses excluded; share normalized to valid responses per AI type.
Source: Assemble’s Decision Intelligence Benchmark — Q1 2026

Actual AI Use

Actual daily use is lower than expectations across all three categories, with the most prominent gap in enterprise AI. Senior leaders report higher actual team use of public GenAI (56%) and enterprise AI (44%) than directors (43% and 17%, respectively).

For company-owned AI, actual observed usage is the same across both levels — 33% — which means senior leaders and directors are seeing their teams use the tools equally, even as their expectations remain dramatically different. Members describe situations where the tools are in place, but the workflows and habits haven't yet formed around them.

Actual Daily Use by AI Type
Q: What is the actual frequency of use from your team for each AI type?
(N=17) · "N/A" responses excluded; share normalized to valid responses per AI type.
Source: Assemble’s Decision Intelligence Benchmark — Q1 2026

The Gap

Across all three AI types, the gap between expectations and actual use is significant. Public GenAI shows a 14-point gap, company-owned AI shows a 17-point gap, and enterprise platform AI shows the widest divergence with a 34-point gap. Enterprise AI requires system integration, workflow design, and a user base confident enough to make these tools part of how they work — these may be opportunities for supply chain leaders.

The divergence on company-owned AI expectations between senior leaders and directors makes that figure almost misleading: senior leaders' high expectations aren't softening even as directors see little day-to-day use. Closing that expectations gap may be as important as closing the adoption gap.

The Gap: Expected vs. Actual Daily AI Use
Percentage-point gap between expected and actual daily use, by AI type
(N=17)
Source: Assemble’s Decision Intelligence Benchmark — Q1 2026
Our Take

Supply chain leaders are not setting moderate targets. They're running adoption expectations on the higher end compared to peer functions. Supply chain is one of the functions ripe for automation and "always-on" monitoring to help avoid shipping or logistics issues. As leaders increasingly manage cost, service, competitive position, and regulatory complexity, adopting new processes can be difficult. The gap on enterprise AI in particular points to a function where the tools have been procured, but the operational habits haven't formed. Members describe a similar phenomenon at the senior leadership level: Executives believe the tools are in place and delivering, but the teams managing day-to-day operations are still behind on adoption.

Three actions for leaders
Part 2: AI Dependence & Automation Outlook

Supply Chain Sees More Automation Ahead — and Is Not (Yet) Significantly Dependent on AI

Supply chain leaders hold a more expansive view of AI automation potential than most peer functions. But their current dependence level tells a different story — one where AI is not yet fully embedded for maximum impact.

Highlights from the data
  • Supply chain optimistic on automation: 65% of supply chain leaders expect more than 20% of their work to be automated within two years. The function has some of the highest expectations of automation, with the highest share of responses in the 21-40% range and the second-highest share of the 11 functions in the 41-60% range.
  • Not too far, though: None of the supply chain leaders expect over 80% of their work to be automated in 24 months, while four other functions had at least a few who felt that way. Supply chain's automation optimism is clustered in the 21-60% range.
  • Dependence on AI is currently moderate: 24% report they would face moderate disruption if AI disappeared tomorrow — near the 27% cross-functional benchmark. None of the leaders said losing AI would be a major disruption, and the highest share (41%) said it would only be a slight disruption.

Dependence on AI

Supply chain leaders are not yet reporting the kind of AI dependence that suggests deep integration. Twenty-four percent say they would face moderate disruption if AI became unavailable, with 41% reporting that they would face slight disruption, and 35% saying operations would continue as normal. Supply chain runs slightly below the cross-functional benchmark, where 27% report moderate or greater disruption.

That relative restraint is not what members describe on calls or the expectations for usage. The mismatch may reflect a distinction between AI that teams have in place and use daily versus AI that is still in the evaluation and implementation phase. Supply chain functions carry large tech stacks and active expansion plans, but actual daily use of emerging tech can be a slow-moving process. It looks like most supply chain teams are working to get there but have a long way to go.

Disruption if AI Disappeared Tomorrow
Q: If AI were suddenly unavailable tomorrow, how disrupted would your function be?
(N=17)
Source: Assemble’s Decision Intelligence Benchmark — Q1 2026

AI Automation Outlook

Supply chain leaders stand out in their expectations for their work to be automated in the next two years. 47% expect AI to automate 21–40% of their work within 24 months, and 18% expect 41–60% automation, the first and second-highest figures among functions in those ranges, respectively.

As a technology-heavy function, supply chain produces a lot of data and contains a lot of opportunities for technology to support manufacturing, shipping, and logistics organizations by automating or monitoring various processes.

Percent of Supply Chain Work Automatable by AI in the Next 24 Months
Q: Approximately what percentage of your function's work could be automated by AI over the next 24 months?
(N=17) · "N/A" responses excluded; distribution normalized to valid responses.
Source: Assemble’s Decision Intelligence Benchmark — Q1 2026
Our Take

Supply chain leaders see the potential of an organization supported by AI. They expect it to reach further into operations faster than almost any peer function. But current dependence and usage has not yet caught up to that outlook; the gap between anticipated automation and actual daily embedded use is significant. Leaders need to get creative in remaking workflows to adapt to the opportunities afforded by emerging technology.

Three actions for leaders
Part 3: Team Proficiency with AI

Supply Chain Teams Report Modest Proficiency — Especially if You Ask Directors

Across all three AI types, most supply chain teams are rated at beginner or competent. Directors rate their team's enterprise AI proficiency lower than senior leaders, and the data reflects what members describe as a real implementation gap at the working level.

Highlights from the data
  • No experts in public GenAI: Supply chain leaders reported that 56% of their teams are beginners and 44% are competent in public GenAI. They rated employees' proficiency in enterprise AI and company-owned AI at 75% beginner or no experience.
  • Beginner is the dominant rating across all three AI types: 56% of supply chain teams are rated beginner on public GenAI, 69% on enterprise platform AI, and 67% on company-owned AI — the highest beginner concentrations of any of the six AI vectors in this benchmark.
  • Directors rate team enterprise AI proficiency far lower than senior leaders: 100% of directors describe teams as beginner on enterprise platform AI, compared with 50% of senior leaders — a 50-point gap that reflects directors' closer proximity to day-to-day implementation reality.

Proficiency Across AI Types

Supply chain has among the lowest team proficiency ratings in the benchmark across all three AI types. None of the three AI types shows any teams rated expert, with the highest proficiency reported at 13% advanced in enterprise AI and 17% advanced in company-owned AI.

This is not a surprise to those who follow how fast the function's tech stack is moving. Supply chain is adding tools faster than teams can build fluency with them. The data suggests a pressure where executive leadership assumes that new tools mean instant answers, while the teams running those tools are still learning to ask the right questions. Competent teams are using AI. Advanced teams are redesigning work around it.

Team Proficiency by AI Type
Q: How would you rate your team’s overall proficiency in using each AI type?
Beginner / No experience
Competent
Advanced
Expert
(N=17) · "N/A" responses excluded; rows normalized to valid responses per AI type.
Source: Assemble’s Decision Intelligence Benchmark — Q1 2026

The Director Read on Enterprise AI

The most telling role-level divergence in this function is on enterprise platform AI, where 100% of directors rate their teams as beginner — compared with 50% of senior leaders who say the same. Directors are not being pessimistic. They are closer to where the tools actually sit in operations: observing rollout, watching teams struggle to get value from platforms that were purchased with high expectations, and calling what they see.

Our Take

Supply chain is in a challenging position on proficiency compared to its expectations: A function placing unusually high expectations on AI maturity is not at that level. The tools are there, but the human layer — the planners, the analysts, the operations leaders who need to read AI outputs critically, override them when necessary, and integrate them into complex decisions — is still at beginner level. That may be a natural consequence of a function that has invested heavily in technology ahead of the training and enablement for the people who will run it.

Three actions for leaders
Part 4: AI Trust & Mistrust

Supply Chain Trusts Enterprise AI Most

Supply chain shows a trust pattern that rises as AI moves closer to organizational ownership — and two concerns about outputs and data privacy are shared by nearly the entire function, regardless of which AI type is being evaluated.

Highlights from the data
  • Enterprise platform AI earns the highest trust: 65% of supply chain leaders express moderate or significant trust in enterprise AI — above the 51% cross-functional average for this category — and well ahead of the function's 41% trust of public GenAI.
  • Senior leaders trust public GenAI far more than directors do: 50% of senior leaders express moderate or significant trust in public GenAI, compared with 29% of directors — a 21-point gap that likely reflects different exposure to governance risk.
  • Inaccurate outputs and data privacy are near-universal concerns: Both were cited by 82% and 76% of supply chain respondents, respectively. The mistrust was most profound in public GenAI.

Trust Across AI Types

Public generative AI earns the lowest trust: 41% express moderate or significant trust, with the remaining cohort split between neutral and some level of mistrust. Enterprise platform AI performs best — above the cross-functional average — likely reflecting the credibility that comes with vendor accountability and system integration. Interestingly, company-owned AI earns 47% trust, similar to the function's posture on public GenAI.

The seniority cut on public GenAI is the sharpest divergence in the trust data. Senior leaders express 50% moderate or significant trust in public AI tools. Directors come in at 29%. Like proficiency and expectations, the directors' perspective paints a less rosy picture of adoption and success than that of senior leaders.

Trust by AI Type
Q: For each type of AI, what is your level of trust with it?
Significant mistrust
Moderate mistrust
Neutral
Moderate trust
Significant trust
(N=17) · "N/A" responses excluded; rows normalized to valid responses per AI type.
Source: Assemble’s Decision Intelligence Benchmark — Q1 2026

Drivers of Mistrust

The two leading drivers of mistrust in supply chain are inaccurate or hallucinated outputs and data privacy or security concerns. Bias in data or training models, black-box or explainability concerns, and misalignment with internal policy rounded out the major trust drivers. Job security ranks as the lowest concern — notable for a function where AI-driven automation is viewed as both imminent and extensive.

Members describe AI-assisted decisions in contexts where errors carry real cost: a misfired demand forecast, a flawed supplier risk score, or an incorrect inventory recommendation translates directly into service failures or capital misallocation. These drivers of mistrust represent the greatest barriers to improved adoption and ability across teams.

Drivers of AI Mistrust — Public GenAI
Q: When you have trust concerns with Public GenAI, what are the primary reasons? (select all that apply)
(N=17) · Multi-select; denominator is total respondents. Reflects drivers cited for Public GenAI specifically.
Source: Assemble’s Decision Intelligence Benchmark — Q1 2026
Our Take

Supply chain's heavy reliance on data and involvement in thousands of decisions and transactions, including ones where the margin for error is small, means that trust will play a significant role in adoption down the road. Leaders should take these sources of mistrust seriously. Moving to the more trustworthy AI behind enterprise or company-owned systems can help but does not address the core issues. Trust looks likely to build as teams gain hands-on experience — but that trajectory runs through use, not just procurement.

Three actions for leaders
Part 5: How Supply Chain Compares Across Functions

Supply Chain Leads on Ambition — and Trails on the Foundation Beneath It

Compared with peer functions, supply chain stands out on expected AI use, automation outlook, and actual adoption. Where it trails — and substantially — is team proficiency, which is the lowest of any function in the benchmark.

Highlights from the data
  • Supply chain ranks 2nd of 11 functions on expected daily AI use: At 49% average expected daily use across all three AI types, only Data Strategy sets higher expectations — placing supply chain ahead of every other function surveyed.
  • Supply chain ranks 2nd of 11 functions on automation outlook: 18% of supply chain leaders expect 41% or more of the function's work to be automated in the next 24 months — behind only Data Privacy at 19%.
  • Team proficiency is supply chain's significant gap: Only 32% of supply chain teams are rated competent or above across all three AI types, against a cross-functional average of 63% — a 31-point deficit.
Where Supply Chain Stands Out on AI
Supply chain’s distance from the cross-functional average on six AI vectors
Supply chain ahead of average
Supply chain behind average
The benchmark average is the mean of the other ten functions surveyed: Social Media, Healthcare Social Media, ESG & Sustainability, CSR & Social Impact, Talent Marketing, Learning & Development, Employee Experience, Data Privacy, DEI, and Data Strategy. Vectors ordered most to least above cross-functional average: Expected Usage, Automation Outlook, AI Adoption, AI Trust, AI Dependence, Team Proficiency. Hover any bar for supply chain’s score, the benchmark average, and its rank.
Source: Assemble’s Decision Intelligence Benchmark — Q1 2026 (N=17 Supply Chain; 11 functions, cells n≥5).

Cross-Functional Position

Supply chain's benchmark position is defined by a significant gap between ambition and capability. On the vectors that reflect intent — expected AI use (+13 points above average), automation outlook (+10 points), and actual daily adoption (+9 points) — the function leads or nearly leads every peer group. Trust also runs above average (+8 points), driven particularly by strong confidence in enterprise AI.

However, supply chain trails significantly in team proficiency, which sits 31 points below the cross-functional average. No other function in the benchmark shows a gap of this scale between its expectations and capability. That asymmetry represents the function's opportunity to grow alongside emerging technology. Leaders who use this benchmark to make the case for capability investment rather than tool investment will be better positioned than those who treat proficiency as a downstream outcome of buying software.

Our Take

Supply chain's benchmark highlights the function's unique position. It is highly ambitious on AI — expecting the most daily use and projecting the highest automation potential — but also the furthest away from reaching that potential. Investment in training and communication and holding both senior leaders and directors accountable can help raise proficiency. Supply chain is probably ahead of most peer functions in knowing what AI can do. It may need different forms of training and support than other teams like marketing, finance, IT, etc.

Three actions for leaders
  • Present the ambition-capability gap to leadership sponsors. The benchmark data gives supply chain leaders a quantified case: the function is 13 points above average on AI expectations and 31 points below on team proficiency. That gap is the argument for proficiency investment.
  • Use peer function data to calibrate expectations. Supply chain leaders who see where L&D, data strategy, and other functions in their organization stand on proficiency and adoption can use those reference points to set more realistic timelines for their own teams — and more honest expectations with executive stakeholders.
  • Position supply chain's AI data advantage as an organizational resource. Members describe supply chain as leading AI within their organizations because the function holds the data. That position is a platform for broader influence — over AI governance, over data infrastructure investment, and over how other functions build their own AI capabilities.