Logistics Reply, a Reply Group company specializing in innovative solutions for supply chain execution and warehouse management, today released findings from a new survey of supply chain, operations, and technology leaders on how artificial intelligence is actually being used inside warehouses. The research, spanning organizations across retail, food and beverage, consumer packaged goods, manufacturing, third party logistics, and automotive, points to a clear pattern: warehouse AI adoption remains at an early state for many organizations, with nearly half of respondents still evaluating or piloting AI and most organizations using the technology primarily for decision support rather than autonomous action.
The survey found that 27.9% of respondents are evaluating where AI could provide value, while 19.2% are piloting AI in one or two areas. Another 19.2% said AI is being used in one or more warehouse processes, and 10.6% said AI is embedded across much of their warehouse operation. Meanwhile, 23.1% of respondents said their organizations are not currently using AI. These findings suggest that warehouse leaders are approaching AI adoption deliberately, validating business value before expanding into broader deployments.
Where AI is being used today skews heavily toward information rather than action. Reporting and analytics (31.5%) and slotting or replenishment (26.9%) were the two most common use cases, followed by order fulfillment (24.1%) and yard and/or dock management (21.3%). Asked what AI actually does in their warehouse right now, 31.3% of respondents said it generates reports, dashboards, or operational insights, while only 2.5% said AI makes operational decisions autonomously.
People Still Want to Be in the Loop
The survey found a consistent pattern of AI as an advisor rather than a decision-maker. When an operational disruption hits, such as an inventory shortage, a labor gap, or an equipment failure, 47.5% of respondents said AI flags the issue while people coordinate the response manually, and a nearly identical share (47.5%) said AI recommends the next best action while people make the final call. Just 3.8% said AI automatically coordinates the response across systems on its own. Another 1.3% said AI is not involved in operational disruptions.
That caution shows up in how much operations teams trust AI outputs. Most respondents (82.4%) said they usually review AI recommendations before acting on them, and only 5% said they trust AI enough to act without review. When AI does make a recommendation or decision, 74.9% of respondents said employees can override it, either with approval or depending on the process.
Connecting AI Across Systems Is the Next Hurdle
The survey also points to a gap between single-system AI and true cross-functional orchestration. More than half of respondents (58%) said their AI coordinates work across a few connected operational systems, while 35% said it primarily operates within a single application. Only 1% said AI currently orchestrates workflows across multiple systems, such as WMS, ERP, TMS, labor, and automation, together.
That gap lines up with what respondents said they want next. Autonomous decision-making was named the area with the greatest untapped potential (30%), ahead of inventory optimization and labor planning. Looking forward, the top capability leaders want from AI is for it to recommend the best operational actions (23%), closely followed by orchestrating decisions across multiple systems (22%).
What's Paying Off, and What's Holding Teams Back
Respondents pointed to better operational visibility (33.3%) as the single biggest business value AI has delivered so far, followed by reduced manual or repetitive work (25.9%) and improved inventory accuracy and earlier identification of issues or disruptions (19.4% each). But progress is not without friction: budget constraints (25%) and data quality (22.1%) were named the top obstacles to getting more value from AI, ahead of difficulty proving ROI (16.3%), a lack of internal expertise (12.5%), and difficulty identifying the right use cases (11.5%). Integration with existing systems (7.7%), building trust in AI recommendations (3.8%), and change management (1%) were also cited as obstacles.
"Warehouse operators have moved beyond the question of whether AI belongs in the warehouse and into a more important one: how AI can create greater measurable operational value," said Michelle Jones, Director of Presales and Solution Consulting at Logistics Reply. "At this stage, AI is delivering the greatest value by improving operational visibility, supporting better decisions, and reducing manual work. The opportunity ahead is to move from AI that informs teams to AI that can recommend, coordinate, and ultimately orchestrate action across the warehouse."
For more information about this survey data and to explore the key findings in greater detail, please visit https://dam-media.reply.com/co... to download the full report.
