Video game developer Keywords Studios tried to create a game using solely content generated by artificial intelligence but failed because the technology was “unable to replace talent”. The Ireland based developer is one of the biggest outsourcing studios in the world and have worked on several high caliber games such as Baldur’s Gate 3, The Legend of Zelda: Tears of the Kingdom, and Alan Wake 2.
Video games have a particularly complicated and controversial relationship with AI. Embark Studios, the developer of hit shooter The Finals, was criticized for using AI voiceovers by multiple actors and even other developers, as was Naruto x Boruto Ultimate Ninja Storm Connections by CyberConnect2.
As first reported in Game Developer, Keywords said during its latest financial earnings report that it tried using the contentious technology to create a new 2D game solely by using Generative AI. The process lasted six months and according to the developer CEO Bertrand Bodson, it was done in part to determine “where Gen AI has the potential to augment the game development process, and where it lags behind”.
Keywords said the AI tool “identified over 400 tools, evaluating and utilizing those with the best potential” but it “ultimately utilized bench resources [new human workers] from seven different game development studios as part of the project, as the tooling was unable to replace talent”.
“One of the key learnings was that whilst Gen AI may simplify or accelerate certain processes, the best results and quality needed can only be achieved by experts in their field utilising Gen AI.”
There was no info presented on why the experts would need to depend soley on generative AI to make the game, and the product has no planned release. It’s speculated they were trying to cut costs. Keywords laid off many employees last year. The developer is pushing ahead with other Gen AI R&D projects to ensure it can “provide current insights in an ever-evolving part of the market.”
Keywords is still using GenAI tools with its Helpshift division to “aid the customer support journey” in the form of an AI chatbot. The company also recruited AWS (Amazon Web Services) Games’ former head of gaming, Stephen Peacock, to spearhead its AI Centre of Excellence for incorporating GAI.
AVA was intended to be a sci-fi civilization-building game. They presented at the Game Developers Conference (GDC) this year, with Peacock saying: “I think generally the perspective [after finishing Project AVA] was, if I was going to give sort of a star rating for generative AI in general: 3.5 out of 5 for now, not ready…”
We reported on the impact of AI in publishing here, it is still problematic. All these effects in games, publishing, music, etc, were anticipated by professionals, and yet the public relations engines promoting the use of AI continues to convince big companies to invest heavily in hopes of entirely replacing humans.
From the failure of AVA, we can draw the conclusion that the idea of replacing humans with it is a ridiculous pipe dream. Creativity can only come from humans, not pattern recognition engines that have to be reminded how many fingers and toes humans are supposed to have.
So far, the courts have taken the stance that an artificial intelligence, since it is not a person, cannot claim authorship of any produced work. This has a dramatic domino affect on whether or not entire projects that use AI in their creation can be copyrighted, how much claim the studio might have on copyright, and indeed if the use of AI in a project might void the ability of a person or company to copyright a finished work at all.
This raises questions:
- Would content produced by the generative AI tools be owned by the studio using them?
- Would clients accept all risks of games worked on with gen AI?
- Would a studio’s own developers feel ethically averse to using these tools?
- If an employee didn’t use AI, would they be let go?
- If the courts finally rule that using other artist’s work to train AI’s on is intellectual property theft, it will upend the entire industry. What happens then?
Generative AI is already everywhere, and it can be difficult to tell the real from the unreal. There is so much of it that it is crowding out the legimate artists who have spent their lives developing their skills and severely impacting their ability to make a living at their life’s calling. The sheer unfiltered tidal wave of Gen AI content, devoid of human spark or creativity, is the cultural equivalent of grog. It tastes faintly of rum, but it’s mostly water.
The impact of AI video game development is already here. The questions need to be answered by us in law, or they will be decided for us by the big studios. Given the failure of AVA, it seems clear that this isn’t the direction we should be going.
David Raiklen wrote, directed and scored his first film at age 9. He began studying keyboard and composing at age 5. He attended, then taught at UCLA, USC and CalArts. Among his teachers are John Williams and Mel Powel.
He has worked for Fox, Disney and Sprint. David has received numerous awards for his work, including the 2004 American Music Center Award. Dr. Raiklen has composed music and sound design for theater (Death and the Maiden), dance (Russian Ballet), television (Sing Me a Story), cell phone (Spacey Movie), museums (Museum of Tolerance), concert (Violin Sonata ), and film (Appalachian Trail).
His compositions have been performed at the Hollywood Bowl and the first Disney Hall. David Raiken is also host of a successful radio program, Classical Fan Club.
Why does this AI failure not make me feel sad?
“So far, the courts have taken the stance that an artificial intelligence, since it is not a person, cannot claim authorship of any produced work.” Let’s hope that doesn’t change–at least not until AI’s are actually people. I’m thinking of you, Data (and Asimov’s robots).