AI in gaming: Developers worried by generative tech

“I know very well that I could wake up tomorrow and my job will be gone,” Jess Hyland said.

The video game artist says the industry he’s worked in for nearly 15 years is currently in a state of “unrest.”

The surge in players and profits during the pandemic prompted a series of investments, expansions and acquisitions that now seem short-sighted.

While the games industry remains profitable, thousands of workers around the world have lost their jobs and successful studios have closed over the past two years.

There are fears of further closures and cuts.

“We all know someone who’s been made redundant. There’s a lot of uncertainty about the future,” Jess says.

Some executives say generative AI, the technology behind tools such as ChatGPT, could be a savior.

Tech giant Nvidia has shown off prototypes of impressive developer tools, and gaming giants such as Electronic Arts and Ubisoft are also investing in the technology.

With audience expectations rising and budgets for big-budget films in the industry soaring, this seems like the perfect solution.

“The people who are most excited about the creativity that AI will enable aren’t creative people,” says Jess, of the UK Independent Trade Union’s games workers section, who sits on the union’s artificial intelligence working group.

Jess said the widespread layoffs have created suspicions among employees that their bosses are viewing AI as a cost-cutting measure, even though labor is the company’s largest expense.

Jess says she knows someone who has lost their job because of AI, and has heard of others experiencing the same thing.

There are also dozens of reports online suggesting that work on concept art and other traditionally entry-level roles has been affected.

Most of the companies developing AI tools maintain that they are not designed to replace humans, and there is widespread agreement that AI technologies are a long way from replacing humans.

Jess says her bigger worry is that “jobs will change, but not for the better”.

Jess said artists worry they’ll end up supplementing the AI’s efforts, rather than creating their own materials.

For example, publicly available AI image generators can quickly spit out good-looking results from simple text prompts, but are notoriously bad at rendering hands, and sometimes even struggle to render chairs.

“The AI ​​is generating stuff and then you’re going to have to fix it,” Jess says. “That’s not why I started making games.”

Gaming is a multi-billion dollar business, but it’s also an artistic medium that brings together a wide variety of people: artists, musicians, writers, programmers, actors, and more.

A common concern is that AI will minimize creative work rather than enable it.

It’s a sentiment echoed by Chris Knowles, a former senior engine developer at Jagex, the British game company known for its title Runescape.

“If we have to hire actual human artists to modify the output, why not use their creativity to create something new that connects with players?” he says.

Chris, who now runs the UK indie studio Sidequest Ninja, says that in his experience, smaller developers are generally not keen on using generative AI.

One of his concerns concerns cloned games.

Online game stores, where indie developers make the majority of their sales, are flooded with knock-offs of original titles.

Chris said this is especially true for mobile games, where some studios are set up “solely to churn out clones.”

While it’s not yet possible to use AI to rip off an entire game, he says it’s easy to copy artwork and other assets.

“Anything that makes the clone studio business model cheaper and faster makes the difficult task of running a financially sustainable indie studio even harder,” Chris says.

Copyright concerns about generative AI – Currently, some Ongoing litigation – These are currently one of the biggest barriers preventing wider use in games.

The tool is trained on a huge amount of text and images collected from the internet, and like many artists, Jess believes this amounts to “mass copyright infringement.”

Some studios are looking at systems trained on their internal data, and third parties are emerging that promote ethical tools that claim to work from sanctioned sources.

Still, there are fears that AI will be used to produce assets like artwork and 3D models at scale, and workers will be expected to produce more deliverables.

“The more content you create, the more money you can make,” says Jess.

Some in the industry are positive about AI.

“I’m excited to see what AI can bring to music in the near future,” composer Borislav Slavov, who won a BAFTA Games Award for his work on Baldur’s Gate 3, told the BBC.

Speaking at the recent Games Music Festival in London, he said he believed this would enable composers to “exploring musical directions more quickly” and step out of their comfort zone.

“This will allow composers to be inspired and focus more on the essence of composing with deep emotion and strong themes,” he said.

However, he agreed that AI “cannot replace the human soul or mind.”

Jess says that while he has serious personal reservations about using the technology for “creativity automation”, he wouldn’t be opposed to using it to ease the burden of some of the more repetitive administrative tasks common to most projects.

They also need to work to win over another group: gamers.

Online shooter “The Finals” faced backlash over its use of synthetic voices, and developer Square Enix was criticized for its limited use of generated art in multiplayer game “Foamstars.”

Jess believes the increased discussion about AI is making gamers “consider what it is that they love about games and what’s special about sharing experiences that other humans have created.”

“I still put something of myself into it, and I think that’s becoming more and more recognized.”

Indie developer Chris adds: “If you train a generative model only on cave paintings, all you get is cave paintings.

“It takes a human to get from there to the Sistine Chapel.”

Additional reporting by Laura Kress.