We often hear the phrase "Fake it 'til you make it," but when it comes to AI implementation, confidence itself becomes the problem—especially when there's too much of it in the C-suite and not enough on the front lines. A recent Economist Impact survey reveals a troubling pattern: executives and the managers implementing AI agree on the potential for AI in three to four years, but they're worlds apart on what the next 12 months will look like. With 71% of companies accelerating AI implementation, this isn’t a hypothetical problem to worry about. It’s a challenge undermining today's transformation efforts.
Here's the good news: by understanding the specific issues managers and executives cite most often, you can prepare a plan of attack that addresses misalignment before it derails your AI initiatives. We've broken down the reasons for the confidence gap on both sides, so you can start where it matters most.
The timeline disconnect: When will AI actually pay off?
Belief:
A majority of executives say they believe investments in AI will deliver clear financial returns within the next 12 months, but less than half of junior leaders agree.
Reality:
The headlines about AI driving efficiency gains make the math seems simple: invest now, reap rewards quickly. But junior leaders are right to take the long view. AI implementation will take more than a year to show meaningful financial returns for many organizations, just like any digital transformation initiative.
The timeline disconnect is important to understand because it sets up false expectations that can undermine your company’s entire AI initiative. To reframe executives’ thinking, let’s consider what actually needs to happen during AI implementation: selecting the right use cases, integrating with existing systems, training teams on new workflows, gathering sufficient data to train models, testing and refining algorithms, and gradually expanding from pilot programs to full-scale deployment.
Each of these steps takes time, and rushing through them to meet unrealistic deadlines will lead to implementations that fail to deliver benefits. Companies that succeed with AI in their supply chains will be those that allow for the inevitable learning curve and adjustment period.
The possibilities gap: How can we prioritize impactful use cases?
Belief:
C-suite leaders are confident their companies can implement AI across a variety of supply chain functions, from demand forecasting and inventory optimization to supplier risk assessment. But non-executives are twice as likely as their bosses to say it will be difficult to implement a number of AI use cases.
Reality:
To date, only 20% of companies say AI is fully implemented into any of their supply chain planning processes. The good news is that most companies are prioritizing implementations in areas where both executives and managers agree there will be high impact.
The challenge for executives in the short-term will be to balance the workloads of the frontline managers and the teams doing the implementation work. These teams understand the inner workings of your supply chain, like the messy reality of data quality issues and the complexity of integrating new tools with existing workflows. The challenge for these managers is to communicate honestly, early and often, so executives can set realistic timelines and allocate sufficient resources to support AI initiatives.
Together, these groups can combat impossible expectations while delivering transformational results, and lessons from these initial high impact implementations may re-shape what’s possible for other supply chain use cases.
The risk-perception divide: What’s the right balance of techno-optimism and vigilance?
Belief:
Fewer than one in four supply chain leaders consider AI-related risks such as cybersecurity threats, compliance failures, or flawed risk assessments to be major concerns.
Reality:
AI systems in supply chains handle sensitive data about suppliers, customers, pricing, and operations. They make decisions that can have compliance implications, particularly in regulated industries. When AI models make flawed assessments due to biased training data or inappropriate algorithms, the consequences can range from inventory shortages to damaged supplier relationships.
The reality is that AI risks are already making headlines in other industries, and the challenges will not be siloed to sectors outside of supply chain.
Leaders at all levels of the supply chain need to educate themselves about the risks of AI systems, understanding that dismissing these concerns will increase the likelihood and severity of AI-related failures.
The legacy systems challenge: Will technical debt impact implementation?
Belief:
Less than half of C-suite executives (42%) see legacy systems as a critical challenge to AI implementation, even though most of their subordinates express worry about this very issue.
Reality:
Chalk this one up to daily exposure. From the executive perspective, new AI tools should be able to work alongside whatever infrastructure exists. Meanwhile, subordinates have daily experience with outdated systems and processes, like the continued prevalence of spreadsheets. When data lives in hundreds of spreadsheets maintained by different, disconnected functions or is trapped in legacy enterprise systems that can't easily share information, AI projects face enormous obstacles.
And while AI projects may be launched in silos, real benefits will only come from connected, collaborative systems.
The one thing everyone agrees on: Organizational inertia
Belief:
Nearly half of respondents (44% of both executives and non-executives) cite organizational inertia as a significant barrier to using AI for geopolitical resilience and other strategic goals.
Reality:
The challenge with organizational inertia isn't recognizing it exists. It is diagnosing the specific sources of that inertia and addressing them directly. Start from what you know: any concerns about inertia likely stem from past experience with other transformation initiatives that stalled or failed. What can you learn from these events? Is it siloed departments protecting their turf? Is it a risk-averse culture that punishes failures more than it rewards innovation?
All levels of supply chain leaders have optimism and interest in AI. Use this enthusiasm as a jumping off point for hard conversations about the critical issues holding innovation back.
How to build organizational alignment for AI success
The confidence gap between executives and managers around AI implementation in supply chains is real, and it poses genuine risks to transformation efforts. But it's not insurmountable. By acknowledging these differences in perception, having honest conversations about challenges, and aligning around realistic expectations and timelines, organizations can turn this moment of misalignment into an opportunity for building the collaboration and transparency that successful AI implementation requires. With so many organizations reporting acceleration of AI adoption, the time to build that alignment is now.
Learn more about the state of AI in supply chain by downloading the full Economist Impact report, “Supply chain’s big bet on AI for geopolitical resilience” or watch our webinar summarizing the key insights.
Ready to start thinking strategically about your AI deployment? Learn about purpose-built AI for smarter supply chains from Kinaxis.
