Would you trust AI to run your business operations or data center?
Sort by:
I suggest that people read "AI Superpowers". It is a great book that shows the reality of what the future will look like with AI. Also, rewatch the movies Minority Report and Her to see the future, they’re scarily accurate.
As a former CIO and now a board member, I see this as still a fairly nascent technology. From a risk perspective,I would feel more comfortable starting with automating processes that are not mission critical. Just start with auxiliary processes and see how that works, where the cost of an error is not going to kill the enterprise. Then once you prove that some of those less risky processes can be successfully automated, then I will move up the chain to more mission critical processes.
It will be years before we automate the entire enterprise. There is still a lot of human thinking that must be embedded in some of the decisions. I can think of a number of examples, including mergers and acquisitions. There are some intangible things that are very difficult to include in machine learning, but I think we'll have some rudimentary processes automated fairly soon. IPsoft is at the forefront there, but humans still have a role to play in a number of business processes in the enterprise.
It's some time away and I think everybody will do the proof of concept first. Prove to me that it can do this before I start putting it in those mission critical areas. People don't even trust software automation yet. I think there is no easy answer in this space. There isn't anything out there doing full-on AI in the way we would have expected. And because we haven't seen it, we don't know what we were willing to commit to yet. We'll get more comfortable once the data is there.
From a business standpoint, if you think about automation, look at data centers right now. There's still a lot of manual work in data centers because people don't trust software automation, even the basics where we have that automation in so many other layers in the stack.
Everybody's worried about it because of the mission critical nature of the data center. I think it's a psychological thing we have to get over, not believing that it's tried and true, that it's going to work and be consistent, etc. It's going to take a while to get there, unless there's a forcing function, which will be cutting costs. If you need to cut costs, you'll figure out other ways to do the work with fewer people and more automation. AI can give you the ability to scale from one to thousands of support agents immediately because the digital colleagues can handle it. Same kind of thing within the infrastructure. I would trust a machine more than a human, because all of my outages and infrastructure issues have resulted from humans making a mistake. That's just the nature of human cognitive ability: people get tired and may not follow processes, but machines don't do that. I would say I would trust AI more than I would trust the majority of humans that are doing things inside of infrastructure.
When we're talking about AI it really reminds me of the movie WALL-E. Do I want to lose my job and have a robot do my job for me? I feel like there will be an issue with the human component of having a job and doing things, and possibly turning into the people in the movie WALL-E who just sit around, get lazy, and can't walk anymore.
That’s what comes to mind when I think of how far we want AI to go. I don't think we want it to go that far. I want to draw a line that says if people want to sit around and do nothing, fine, they can do that. But if there are other people that want to continue working, they should be able to do that as well.