Industrial AI Podcast

The episode contains several substantive insights about bridging proof-of-concept to production AI in industrial settings, particularly around the gap between consumer and industrial AI expectations, the importance of expectation-setting, and physical asset depreciation cycles as a constraint. However, significant portions consist of self-promotion, meandering discussions about autonomous vehicles and robots that feel tangential, and repeated affirmations rather than novel concepts. The value is moderate but marred by filler.

“The data scientist needs to trigger an update version upgrade for machine replacement. And if he or she knocks on the door of the CFO saying, you know for my model... I really need this machine to be replaced. The CFO goes like Are You Crazy? This Machine is depreciated over seven years Over ten Years and now after four years you want To replace it.”

“The fact that shop floor is about physics and machines impacts the way how that scales in need of physical AI to control them.”

Boris articulates a genuinely useful distinction - that industrial AI requires managing the 98.5% engineering harness around the 1.5% model - and positions this against consumer AI hype cycles. The insight about design space exploration and material discovery via simulation (Siemens Energy, Tesla) is concrete and somewhat fresh. However, the core argument ("industrial AI is different from consumer AI") is not novel by 2024, and the extended riff on autonomous vehicles and humanoid robots as cautionary tales feels recycled. The process mining + agentic AI connection is interesting but underdeveloped.

“Why? because this is digital. You don't need to replace a machine to collect another data point. And this is why the potential of using AI and increasing productivity by AI in product design engineering, it's a lot higher scales better than that actually on the shop floor side of life.”

“The automotive industry spent at least two trillion dollars in the last decade on developing the autonomous vehicle And everyone spent a lot of money. and then if we look at the results, where are we?”

Boris Scharinger brings legitimate practitioner credibility as a Siemens employee with hands-on experience across IT service management, data analytics, and industry 4.0 deployment. He has written a 350-page book on the topic and can reference real internal case studies. However, he is presented primarily as an author promoting his book rather than as an active operator solving live problems at scale right now. His examples (Siemens Energy, Tesla) are well-known and secondhand, not direct execution stories. He is a thoughtful domain expert but not a tier-one operator in active scaling.

“I started my career basically coming from IT management, IT service management which is an interesting place to be because there's already a area where innovation versus stability, right?”

“I did a little bit of work in data analytics for audit and then i moved into the area of industry. four dot zero focused pretty early long before there was this hype and momentum in the space of AI.”

The episode includes some concrete examples (Siemens Energy turbine blades, Tesla giga press, CNC machine idle power consumption, near-shore job losses in Poland/Krakow) and specific metrics (98.5% harness vs. 1.5% model, six-week email approval delays). However, many claims lack numbers: no data on proof-of-concept success rates (mentioned as "very low" with no citation), vague references to "cost factors can be solved" without figures, and broad statements like "We see Use cases in procurement" without specifics. The SAP approver example is concrete but anecdotal and company-anonymized. Overall, specificity is present but inconsistent.

“the gas turbine blade design was so significantly improved by this exercise of design space exploration with the help of simulation and AI, that Siemens Energy became a new performance leader in large gas turbines.”

“ninety eight dot five percent. where the harness what they now call the harness”

The hosts are collegial and appreciative but largely softball in their approach. They ask open-ended book-promotion questions, frequently affirm Boris's points ("Yes," "Exactly," "Right"), and allow him to monologue at length without sharp follow-ups or productive pushback. When disagreement surfaces (e.g., Peter asking if Boris is saying to "forget" shop floor AI), Boris simply restates his position without being pressed. The hosts introduce tangents (mentioning their own discontinued book, the monastery event with Ben Amnuri) that dilute focus. No one challenges the autonomous vehicle tangent, the 2-trillion-dollar claim, or the absence of hard data on POC success rates.

“Boris Welcome To The Podcast. Yay, thanks for having me. I'm so glad to be here and i'm happy too.”

“Perfect! Yes. But before we start can please introduce yourself maybe briefly in two sentences to the listeners?”