Despite billions of dollars spent on supply chain automation, humans are still making the decisions that matter. A new Gartner survey, reported by MHL News, found that by 2030, only 5% of organizations implementing supply chain planning automation will have machines making at least 10% of their planning decisions autonomously. We are in 2026, and the robots are not in charge yet. Not even close.
- 83% of surveyed organizations had spent at least $3 million automating supply chain planning, including AI investments.
- 51% of those organizations spent between $3 million and $10 million on automation alone.
- Gartner projects that by 2030, only 5% of organizations with supply chain automation will reach 10% autonomous decision-making.
- The Gartner survey ran from November 11 through December 18, 2025, across 243 senior leaders globally at companies with revenue above $500 million.
- DHL Supply Chain has deployed more than 8,000 robots across its worldwide operations, according to Supply Chain Management Review.
The stakes here are real. Companies are not spending millions to feel good about their tech stack. They are spending it to reduce error, cut labor costs, and survive the kind of supply chain chaos that COVID-19, geopolitical shocks, and consumer volatility have turned into a permanent condition. If those investments are not producing autonomous decisions, someone needs to answer for the gap.
Money Does Not Buy AI Readiness
Buse Aras, director analyst in Gartner’s supply chain practice, put it plainly: “Spending millions on supply chain planning automation can expand technical capabilities, but investment alone does not create AI readiness.” That sentence should be tattooed on the wall of every boardroom that approved a nine-figure technology budget based on a vendor’s demo reel.
The actual bottleneck is not the software. It is the people, the data quality, and the organizational structure that the software has to work inside of. AI cannot make smart replenishment decisions when the underlying inventory data is three weeks stale and the team responsible for maintaining it is underwater. The tech is only as smart as the inputs it receives and the processes wrapped around it.
What is genuinely frustrating is that this lesson keeps getting learned the hard way. The pattern is familiar: a company writes a big check, deploys a shiny planning platform, and then discovers that their data governance is a disaster, their planners do not trust the system’s recommendations, and middle management is quietly overriding the AI outputs anyway. The technology sits there humming, and a human with a spreadsheet is still calling the shots. That is not a failure of AI. That is a failure of implementation discipline.
This same pattern is playing out in other sectors too. Construction robots are tackling America’s housing shortage, but we’re a long way from humanoid homebuilders — and the reason is identical: the physical and organizational infrastructure required to hand real decisions to machines is far harder to build than the machines themselves.
The Human-Machine Balance Is Not a Temporary Problem
Here is the contrarian read that nobody in the automation vendor community wants to say out loud: maybe the 5% figure by 2030 is not a failure. Maybe it is the correct outcome.
Supply chain planning is not a solved problem. It involves geopolitical risk, supplier relationships, demand signals that shift overnight, and judgment calls about trade-offs that carry real-world consequences for real people. Replenishment and order prioritization, yes, those can be automated with meaningful confidence. But strategic decisions about where to source, how to respond to a port closure, or whether to hold inventory ahead of a tariff hike? Those decisions benefit enormously from a human who carries context that no model has been trained on.
Matt Gregory of Infios described it well: companies need to sense what is happening across their supply chain, make better decisions, and turn those decisions into action. That framing is telling. The tech helps you sense and act. The decision still belongs to someone.
The logistics operation of the future will not be fully automated nor fully human — it will be both, as Supply Chain Management Review put it in its analysis of physical AI and robotics deployment. That is not a compromise. That is the architecture. The question is whether companies are building toward that hybrid with intention or just throwing money at tools and hoping the org chart catches up.
The same dynamic is visible across industries where technology and power intersect. Technology billionaires are rewriting world politics precisely because they understand that the people who control the infrastructure of decisions — not just the tools — hold the real leverage. Supply chain is no different.
And consider where this logic leads in healthcare: AI is personalizing medicine, but nobody is suggesting we cut physicians out of treatment decisions yet. The pattern holds. AI accelerates. Humans decide. The industries that pretend otherwise are the ones that end up on the wrong side of an audit.
The companies that will pull ahead by 2030 are not the ones that spent the most on automation — they are the ones that spent it in the right order, fixing their data foundations, retraining their planners, and building trust between the humans and the systems before scaling anything. Watch which supply chain leaders start talking about decision architecture rather than just technology spend; that is your signal that someone actually gets it.
Watch the Breakdown
Sources
- Supply Chain Planning Decision Still Made by Humans — www.mhlnews.com
- Exploring Supply Chain Automation and Growth at Infios — supplychaindigital.com
- Beyond the hype: Building flexible and scalable supply chains in a VUCA world — www.scmr.com
