You can't slow down AI, instead focus should be on strengthening security, defences: Cisco CEO

Cisco Chairman and CEO Chuck Robbins on Wednesday said any effort to slow down AI is unlikely to work, and that the focus should, instead, be on strengthening security and defences to keep up with the technology.

Amid global calls for pause in Artificial Intelligence and scientists warning about possible catastrophic outcomes for humanity from rogue AI, Robbins, in an interview with PTI, said, "I think we are going to be okay".

"I think the biggest thing...there is this whole debate about should we slow down AI or not. You are not going to slow it down. I mean, I just don't think that is going to happen," Robbins said.


Counter defences, he said, have to move "at the same pace or faster" than AI.

"So what we need to make sure we are doing is that we actually drive the security side of it, and the defensive side of it, and all the things that we need to do to prepare for the speed at which AI is moving...We have to move at the same pace or faster, and I think that is what we need to focus on," he said.

Cisco's top leader further said there should be logic around how models are released and around guardrails, and that governments and companies should know what is going on.

"We certainly need to have logic around how models get released, logic around guardrails. We should know what's going on. But every technology we've ever introduced has had positives and negatives. I mean, you just have to solve for the negatives. And as I've said, if it's a computer science problem, I know we can solve it, and I think with enough technology, we can prepare for some of the cyber risk and other things that we are concerned about," he said.

On whether regulations are imminent, Robbins said rules are followed by people with integrity, while those looking to attack systems will ignore them.

He said the more pressing risk today is that AI models are exposing gaps in basic cyber "hygiene" at customers, such as password management, patch management and the use of technology that is past its last day of support.

"I think guardrails, sure...I think understanding the risks, absolutely. But I think we just have to focus on the security side of it. Focus on effective guardrails, focus on hygiene," he said.

Robbins said the biggest task for India, and indeed any country, is balancing the pace of innovation with appropriate regulation. If regulation slows innovation, he said, a country falls behind.

"I think the biggest thing is balancing the speed and the pace of innovation with the desire to put in the appropriate regulation. I think getting that balance right is not just for India, I think it is going to be true for any country that wants to be successful," he said.

Robbins added: "I think you are going to have to make sure you get that balance just right, so that your regulation doesn't slow down the innovation. Because if it does, you fall behind. If you don't have the appropriate regulation and whatever that looks like...but the tendency is to go too far one way, and I think finding that balance is important."

On jobs, he said he does not see AI as a net job disruptor and expects it to move people up the value chain.

"You can either do the same amount of work you're doing today potentially with fewer people, and have the same output, or you can decide you're going to keep the number of people you have, redeploy them and move them into other spaces, and leverage AI effectively to fundamentally increase your output at the same cost, which is better for your P&L. It's better for your customers. It's going to get more innovation," Robbins said.

The winners will be companies who choose the path of leveraging AI to optimise value, versus those who choose it merely as a cost saving tool, Robbins said.