The designer of Atari's notorious E.T. likens the deluge of AI games to events that preceded the '80s industry crash: 'There was a glut of crap on the market'

Image of the box art for Atari's E.T.: The Extra Terrestrial
(Image credit: Atari)
0

The emergence of generative AI coding tools has numerous effects on the games industry this year, but the prominent of these is that it's made Steam an absolute nightmare to navigate. Since the summer, Steam (and other storefronts like itch.io) have been inundated with clones and AI slop games, as AI users turn to tools like Claude to output thousands of low-quality titles.

For many of us, this seems like a new problem. But for one veteran game developer, the scenario is all too familiar. Howard Scott Warshaw is the developer of numerous Atari 2600 games including Yars' Revenge and Raiders of the Lost Ark. But perhaps his most famous (or infamous) game is 1982's E.T. the Extra Terrestrial, the notorious tie-in game that was so terrible Atari buried unsold copies of it in the desert.

E.T. has been cited as a contributing factor in the US video game industry crash of 1983—one of many poor-quality 8-bit titles that hollowed out the gaming market in the States. It's here where Warshaw sees a parallel to today's emerging scenario, though he doesn't necessarily believe it will end in the same way.

Latest Videos FromPC Gamer

In a talk he gave at the DevGAMM Lisbon conference last week (via GamesIndustry.biz) Warshaw remarked upon how AI coding tools have seemed to rapidly sped up iteration times on game development. "People who used to spend weeks, months or years programming something can now do it in days with AI support," he explained.

"So what does this mean in game development? What it means is we're going to see a lot more games produced. You ever hear of the crash of '83 or '84, the big video game crash? They said the problem that happened in the crash, and I think that was true, was there was a glut of crap on the market, because the machine opened up and anybody could just pop a game out, and they did."

Warshaw pointed out that, at the time "good games were still being made" but these games were "way overshadowed by the crap: and that created a real problem in the game industry." He sees a similar situation emerging now thanks to generative AI tools. "AI is going to have the same impact on a larger scale, because on one level, so many more people can now realise and create a video game than ever could before just using AI. And that's okay, but there's going to be a lot of crap, because most games suck. That's just the way it is."

The reason for this, Warshaw says, is that, like all artistic endeavours "there's some things in life that everybody can see themselves doing, and the truth is most people who do it can't do it very well. And what AI is going to do is make that more possible for people who can't do it to do it." Yet while he thinks that most of the games that result from these ideas "aren't going to be very interesting" he also believes that "it's also going to enable the people who really do have some innovative, interesting and solid ideas to realise those, too."

In short, Warshaw believes that AI will provoke a shift away from technology development to a greater focus on game design, which he ultimately believes is "actually going to be a super positive thing for the video game industry." He also said that, despite observing similar warning signs as in 1983, he doesn't "see the prospect of a crash in the video game industry" and instead expects to see a "reorganisation, where we need better filters to figure out a way to get rid of the garbage games".

Warshaw is certainly correct about AI tools resulting in a flood of crap, but I also think he himself overvalues the inherent worth of ideas in terms of what makes a great game. While ideas themselves can be inherently good or bad, no idea emerges from the creative process unscathed. Indeed, a major factor in creating art of any kind is engaging and grappling with that process, learning how to deal with the friction between the image in your head and what ends up on the page, or the canvas, or your scripting program. That mastery only comes through practice and experience. It cannot be outsourced or circumvented.

But that doesn't change the fact that storefronts are currently being overwhelmed with terrible, AI-assisted games, and it's having all manner of knock-on effects. Earlier this month, Valve admitted it is "running behind" on approving Steam games because it has to keep re-reviewing them, and although Valve didn't cite AI as the cause of this, it's an awful coincidence that this has started happening as AI coding tools result in game submissions shooting through the roof.

2026 gamesBest PC gamesFree PC gamesBest FPS gamesBest RPGsBest co-op games

2026 games: All the upcoming games
Best PC games: Our all-time favorites
Free PC games: Freebie fest
Best FPS games: Finest gunplay
Best RPGs: Grand adventures
Best co-op games: Better together

Contributor

Rick has been fascinated by PC gaming since he was seven years old, when he used to sneak into his dad's home office for covert sessions of Doom. He grew up on a diet of similarly unsuitable games, with favourites including Quake, Thief, Half-Life and Deus Ex. Between 2013 and 2022, Rick was games editor of Custom PC magazine and associated website bit-tech.net. But he's always kept one foot in freelance games journalism, writing for publications like Edge, Eurogamer, the Guardian and, naturally, PC Gamer. While he'll play anything that can be controlled with a keyboard and mouse, he has a particular passion for first-person shooters and immersive sims.

You must confirm your public display name before commenting

Please logout and then login again, you will then be prompted to enter your display name.