CIO Leadership Live

There are a handful of genuinely useful observations - particularly the AI model surfacing a customer-service root cause when the team was querying for a pricing problem, and the 15-minute delivery benchmark - but these are surrounded by significant padding, generic transformation narrative, and well-worn cybersecurity platitudes that add little value for a B2B operator.

“The answer was right, the question was wrong.”

“we have been able to automate the ordering process from a single click at a hospital nurse station to our branch operations where we get all the information we need to deliver life-saving equipment. And in one hospital we actually deliver a full bed with a mattress or surface as we call them, ready to go in less than 15 minutes”

The anecdote about misframing an AI query and discovering the model returned a more accurate answer than the question deserved is a genuinely non-obvious lesson; everything else - SaaS migration, garbage-in/garbage-out, humans as the weakest security link - is recycled industry boilerplate.

“we felt we have a pricing issue. So we focused all our questions about pricing. When we fed the information to the model, the model came back when it was a customer service issue.”

“we have a specific email in the company where if any employee thinks there's any risk, they can send what they think is a risk to that particular email... every quarter we raffle an iPad.”

Marin is a working CIO at a niche but operationally complex healthcare logistics company, clearly describing real implementations rather than theory; he is not a household name or scale operator, but he is a genuine practitioner with hands-on experience, which keeps the score solidly mid-range.

“if you can get all the equipment that we have rented in the last 23 years, which we know the data, we have the data and start feeding those models, then we can actually be more predictive than sometimes probably some manufacturers would.”

“We create a system that we call GoUSME Connect. It's a combination of RPA, AI and machine learning.”

A few concrete numbers land well - 15-minute delivery, 2-hour SLA, 12-24 hour industry standard, 23 years of rental data - but the episode largely stops there; no cost figures, adoption rates, error-reduction metrics, or scale data are offered, leaving many claims unsubstantiated.

“in one hospital we actually deliver a full bed with a mattress or surface as we call them, ready to go in less than 15 minutes”

“our SLAs are two hours plus driving time. Industry standards normally are between 12 and 24 hours.”

The host does ask a few good follow-up probes ('What did it used to take you?', 'Can you compare the two?') that yield the episode's best data points, but the questions are mostly predictable conference-interview fare, there is no pushback on vague claims, and the final question devolves into gadget small-talk.

“What did it used to take you?”

“Can you compare the two? How did you do it and how do you do it now?”