Elon Musk said it's probably unwise to kick robots because they have a very good memory

Scott Olson / Getty Images

Elon Musk, the CEO of Tesla.

• Elon Musk said it is "probably not wise" to kick robots in a wide-ranging interview with Joe Rogan.
• Musk, who has famously voiced concerns about AI, said he worries less about it now because he has adopted a more fatalistic attitude.
• He believes the danger with AI will come when people weaponise it against each other.

Robots are a bit like elephants, they never forget. That's according to Elon Musk at least.

Musk expanded on previous remarks about AI during two-and-a-half-hour long podcast the "Joe Rogan Experience," presented by comedian and former UFC commentator Joe Rogan.

Complimentary Tech Event

Transform talent with learning that works

Capability development is critical for businesses who want to push the envelope of innovation.Discover how business leaders are strategizing around building talent capabilities and empowering employee transformation.Know More

Rogan joked about animal protection group PETA putting out a statement that you shouldn't kick robots, in reference to footage of Boston Dynamics robots being kicked around by engineers to test their sturdiness.

And engineers are not the only culprits, Starship Technologies' food-delivery robots are sometimes kicked by members of the public, the company's cofounder told Business Insider in June.

"It's probably not wise," said Musk, as Rogan laughed. "Their memory is very good."

Musk's fears on the subject of AI are well documented, but he told Rogan that he worries about it less these days because he has adopted a "fatalistic attitude."

"It's not necessarily bad, it's just it's definitely going to be outside of human control," he said. According to Musk the big danger going forward will be that people will weaponize AI, and in July he signed a pledge along with various other tech leaders never to develop lethal autonomous weapons.

AI has already been weaponised to an extent, and has military applications such as targeting software in drones. It also has vast potential for weaponisation in cyberspace, and an Oxford professor told Business Insider that this is where the real danger lies, as it could autonomously escalate an international conflict.

You can watch the full Elon Musk interview here: