Sr. Director, Security Engineering & Threat Management, 10,001+ employees
GE realized at one point that there was so much industrial data that there was value to be had from taking a step back. The revenue model changed at one point where they realized that it wasn't just about production and construction of things, but the service that came along with them. And that service model allowed this next industrial data model to kind of proliferate from it software-wise. So taking that industrial data and then doing modeling and things with it. So a lot of the AI and ML based analytics that we develop tend to be very industrial oriented. There's no shortage of new ways to create an analytic, to see when something's about to have a failure. And the reason that's important, for people who don't really have a contextual background in industrial assets, is we're talking at a scale of things that is just not really normal to folks that work in a cube and make a PR to get help every now and then. It's completely different. Like when a turbine goes down, you might be talking like tens of hundreds of millions of dollars: the downtime, the maintenance, the average window. If you can pay a few million dollars on an annual revenue cycle to monitor that and avoid that, the breakeven is so not even a forethought that you just go and do it.There's so much value in that. That's really where a lot of the AI and the work in this from a data science perspective happens. There's such an opportunity for these products to see and detect patterns that you just simply can't. The immediate value prop I see is you're able to look through more data than anybody I have can at scale.CISO in Software, 51 - 200 employees
My last company was a manufacturing company. Our manufacturing lines and equipment to produce our product is super expensive, and we had instances where that line would go down and then in our CMOs, it would take us three or four days to get that line back up and manufacturing again. So that was a big problem. What we were trying to do is to predict when these machines would need maintenance or when they would go down based on some historic model, so that we could easily shift from one line to another without having any downtime. But since these were new machines, it was difficult to do with traditional modelling.Chief Digital Officer, 201 - 500 employees
The business that I manage is beauty products and we're trying to anticipate our consumer’s needs, and skincare is a big need. It’s a product that we're trying to grow. We have a loyalty program where we have high value customers and we know that they love our skincare products, so we really use that to give our consumer service group data around, “Hey, they bought this six months ago. It's about to run out.” How do you incentivize them or reach out to them in a way that will get them to buy stuff. My use case there is a little bit different, it's anticipating the consumer's need, even before they actually know they need something.Content you might like
Yes, significantly26%
Yes, but not by much59%
No13%
Not sure2%
82 PARTICIPANTS
Big Data21%
Remote Work17%
Microservices / Containerization11%
CI / CD5%
Zero-Trust15%
Automation2%
Digital Transformation16%
Cloud / Cloud Native1%
DevOps or DevSecOps6%
Other (comment)1%
1006 PARTICIPANTS
CTO in Software, 11 - 50 employees
No, we haven't published corporate guidance establishing guardrails for use of commercial generative AI services.
What I hear so often from these new products coming out is well, we downloaded it and it found all these things we didn't know about, so we had to buy it. Well, does that mean that you're not necessarily doing your job that great or the product is that great. This leaves a big gray area. One of the things that I always felt really interesting about what Anil was doing with this product is, he's not trying to reinvent something someone's already done. There is no shortage of data. There's no shortage of data aggregation. Everyone's already pulling data in all sorts of ways. They're applying this really compelling logic structure to what you already have, and tapping into that value to bring you additional value. They're saying, we've got these really solid use cases of what we try and detect on within your own data streams or data sets. We want to help further enable what you already have. I think that's something that's top of mind for me.