Why AI Pilots Fail in Manufacturing (and How to Fix It) with Russell Halper
AI pilots look great in a controlled environment. Then they hit the real shop floor and quietly fall apart. In this episode, Lisa Ryan talks with Russell Halper, founder and managing director of Insight Kitchen, about why so many manufacturing AI initiatives stall after deployment and what leaders can do differently before they ever write the first prompt.
Russell's path here is unusual. He started a PhD in mathematics under a founder of chaos theory, moved into operations research (one of the earliest forms of data science), worked at UPS, then went in-house as an internal consultant for a major consumer goods manufacturer. From there, he spent eight years at a boutique Palo Alto consulting firm, helped grow it from 16 to 100 people, saw it acquired by a large global consultancy, and ran its West Coast Generative AI practice. Two years ago, he left to found Insight Kitchen, a boutique data science and AI consultancy focused on operational AI in manufacturing and supply chain.
Key Takeaways for Manufacturing Leaders
A pilot is not a proof point; it is a rehearsal for scale. Most organizations design pilots to prove AI works in a controlled setting. Russell argues the real work is designing the pilot to surface the change management, data, and complexity issues that will show up at full scale, before you ever roll it out plant-wide.
A 95 percent solution can be a catastrophic failure. In operational environments, the last few percentage points of accuracy often carry the most real-world risk. Before adopting an AI tool, define what "good enough" actually means for that specific decision, because the threshold changes by use case.
Watch for compounding error, not just error rate. If an AI model is 99 percent as good as your best people but making decisions much faster, small errors can compound quickly. Ask whether faster, slightly less accurate decisions are actually creating value or just creating a bigger mess sooner.
Usage is your earliest warning sign. If the people on the floor are not acting on the AI's recommendations, you already have your answer. A model nobody trusts is not a decision-making tool, no matter how sophisticated it is.
Your senior operators know things your data never captured. A 40-year machine operator who says "something's off with this machine" is picking up on signals that never made it into a sensor feed. When leadership overrides that instinct because "the dashboard is green," they lose the tribal knowledge as retiring workers walk out the door.
"Clean data first" can be a stall tactic. Russell has seen organizations spend endless cycles perfecting data before doing anything with it. His advice: work the actual business problem and pull the data along with you. You will often learn more about your data by using it than by polishing it in isolation.
Start with the business problem, not the technology. The mechanics of building a financial case for an AI project look almost identical to the case for any other software investment. Start with the outcome you want (efficiency, reduced reject rate, faster throughput) and work backward to whether AI is the right tool, not the other way around.
Mid-market manufacturers may have a real speed advantage right now. Fewer legacy systems, more focused operations, and lower internal liability friction mean mid-sized manufacturers can often move faster on AI adoption than large enterprises, which are weighed down by legacy tech and internal politics.
As AI gets better, ask whether your business is getting better with it. Russell's framework: track whether your organization's capability is compounding alongside the technology, or whether you are just layering new tools onto the same old process.
Discussion Questions for Your Leadership Team
- Where are we chasing a "bright, shiny" AI tool instead of a defined business outcome?
- What does our frontline team know that never makes it into our systems, and how are we capturing it before they retire?
- If usage of a tool dropped to zero tomorrow, would we notice, and would it matter?
About Russell Halper
Russell Halper is the founder and managing director of Insight Kitchen, a boutique data science and AI consultancy that embeds AI directly into the processes that run supply chain, manufacturing, and revenue management operations. He holds a PhD in Applied Mathematics and brings over 20 years of experience at the intersection of applied mathematics, operations research, and AI, having advised startups, private equity-backed SaaS companies, and Fortune 500 organizations across retail, consumer goods, manufacturing, high-tech, and logistics. Before founding Insight Kitchen, Russell was a partner at End-to-End Analytics (acquired by Accenture) and later led Accenture's West Coast Generative AI practice as Managing Director.
Connect with Russell:
- Insight Kitchen: insightkitchen.ai
- LinkedIn: linkedin.com/in/russellhalper
About the Host
Lisa Ryan, CSP, MBA, is the founder of Grategy® and Chief Appreciation Strategist. She hosts the Manufacturers Network Podcast and speaks to manufacturing, industrial, skilled trades, and healthcare audiences on employee retention, workplace culture, and the human side of AI and automation. Her newest book, Smart Plant: Aligning AI, Automation, and People, explores how manufacturers can adopt AI without losing the judgment and tribal knowledge their people bring to the floor.
Book Lisa Ryan for your next event, conference, or leadership training on AI, culture, and workforce retention:
- Website: LisaRyanSpeaks.com
- Email: lisa@grategy.com
Connect and Subscribe
- Listen to more episodes of the Manufacturers Network Podcast
- Learn more about Grategy at LisaRyanSpeaks.com
- Get Smart Plant: Aligning AI, Automation, and People: amazon.com/dp/B0GZMFBVZ2
