Randy Pitchford says Gearbox has a firm no-AI policy after posting an AI-generated image of an AI working at Gearbox

Gearbox boss Randy Pitchford is making friends and influencing people on social media, which is to say that he’s repeatedly insisting the studio does not use generative AI in a professional capacity (and occasionally calling people names) after a gen AI image he posted over the weekend drew a pretty negative reaction from followers.

The whole thing started on May 3: “I asked my primary AI tool to generate a selfie that indicates how they feel based on how I interact with it and this is what it generated (note: background words were not prompted and have zero relationship to anything real),” he wrote on X.

(Image credit: Randy Pitchford (Twitter))

Boy, that’s a ratio.

In a reply to a follower who said a real photo of Pitchford in his office “would have been much better received,” he wrote,” The point was to make the AI make a picture of how it saw itself to put a lens on the absurdity of the idea of AI having an identity. In what universe does what you suggest have anything to do with the point?”

But it was all downhill from there: Other followers expressed concern about the use of generative AI in Borderlands games, or disapproval in general; Pitchford insisted several times that the intent was merely to be “silly,” and that Gearbox’s policy “is that we do not use AI for anything in any professional capacity that any customer could ever see.”

(Image credit: Randy Pitchford (Twitter))

(Image credit: Randy Pitchford (Twitter))

(Image credit: Randy Pitchford (Twitter))

(Image credit: Randy Pitchford (Twitter))

The pump for this particular backlash may have been primed by recent Borderlands 4 patch notes, which some players suspected were written or modified by AI. Pitchford insisted that’s not the case, however, saying in response to one such accusation that “errors in patch notes are human error.”

Eventually, Pitchford posted a longer statement saying he was inspired by friends who were “goofing around with making [AI] try to make pictures of itself.”

“My friends I was with for lunch earlier produced some funny things with similar prompts for themselves and I wanted to see what bullshit it would generate as an idea of a self identity because the idea of an AI even having an identity is nonsense,” Pitchford wrote. “The result was somehow more embarrassingly hilarious than I expected and I wanted to share that.

“ChatGPT has no information from me about anything from my work because I don’t use AI for work and our policy is no AI in any work that could ever be seen by any customer. I’m using my personal phone and not my work computer (which is isolated from personal systems). It got whatever it generated from whatever public knowledge of Gearbox it has access to (hence my very clear disclaimer) and the timing or content of this has exactly zero to do with whatever feelings you’ve spun yourself up about with patch notes.”

He then urged followers to “maybe relax a little and have some fun?”

That’s probably not a bad idea, generally speaking, and honestly I don’t doubt that Pitchford was just “goofing around,” as he put it. At the same time, I feel like a little contrition would’ve gone a long way here. As we’ve seen many times in the past, the use of generative AI in game development is a very touchy topic, and I don’t think it’s at all surprising that a high-profile game industry executive posting gen AI images bearing the Gearbox logo and various other game dev trappings would elicit a strong reaction, and not necessary of the positive sort. Games industry executives, and more importantly the money guys, may be eager to embrace it, but people who actually make and play games? Not so much.