To celebrate Polygon’s 10th anniversary, we’re releasing a special edition: Next 10, to consider what games and entertainment will be from some of our favorite artists and writers in the next decade. over here, Far cry 2 And Watch Dogs: Legion Creative Director Clint Hocking looks at the future of AI-generated games.
In the year In 2032, we humans will rise above our base minds and the world will finally be at peace. There will be no more crime, no more violence, no more discord. Because there are no drugs or alcohol, we will be protected from dangerous toxins, and global warming and related problems will be forgotten for a long time. Everything is perfect until a dangerous supervillain named Phoenix escapes from cryo prison and we have no choice but to release a powerful and underpowered cop named Spartan from a screaming prison (don’t worry, he was innocent all this time) to save us all.
Of course, this does not look like 2032 – this is the plot of the 1993 film. A demolition man, starring Sylvester Stallone, Wesley Snipes and Sandra Bullock and directed by Marco Brambilla. Unsurprisingly, the movie’s 2032 looks nothing like what the actual 2032 will be like, although to be fair to the creators, that probably wasn’t their goal. In the year In 1993, it would have been unthinkable for someone sitting on a bus today to tell the phone to play. A demolition manAnd within five seconds, the digitized film is streamed from the data center over cellular wireless networks directly to the 5.8-inch HD touchscreen.
Although this fact may seem banal and almost insignificant to us, it is difficult to imagine how much the technology chain has defined the structure of the world in 2022. Networks, content digitization, data compression and streaming, recommendation algorithms, machine learning, trillion dollar corporations, media and telecommunications empires, and supercomputers/malls in everyone’s pocket – all these things have been abolished under neoliberal compulsion. Let’s keep our crumbling post-industrial economy on life support by reconciling our entire species with digital entertainment. So where does this all lead?
In the year A person sitting on a 2032 bus can tell their phone to play. A demolition man, written and directed by Jordan Peele and starring Millie Bobby Brown as Spartan, Lil Nas X as Phoenix and Tom Holland as Sandra Bullock. Machine learning can see not only how to rewrite the 1993 script to make it more unsettling and strange, but also how to shape and illuminate individual scenes — simultaneously analyzing every word Jordan Peele wrote. and bullets. 3D sets and locations are created, the stars are aged to match the script’s requirements, and their performances and voices are all blended together, with license fees instantly sent to their bank accounts. The entire process of creating a two-hour movie from a 40-year-old script and the performances of everyone involved takes several minutes, but with today’s streaming services, you can start watching. In a few seconds, when the film flies in the clouds. The technologies that enable this future are mostly already there—yet not robust or well integrated. I believe that today we are close to the future, and the technological base is well established, it is not a question of whether it will happen. It’s just a question of when. But the main question is, how will this new future change us and our culture?
As a game developer, I sometimes feel like I have a front-row seat to this change. I have already produced works that can be radically experienced by a variety of audience members. This has always been the case with games; No two football games or walks are the same. In recent years, there have been many more attempts to present this unique experience in stories as well. in Watch Dogs: LegionWe set out to make a game where we deliberately stopped taking control to the player: you can be whoever you want, and when you watch a cutscene, we don’t know who’s going to be in it. Just like the original Pele version A demolition man It still has the same general plot as the 40-year-old original, we know what each scene is about and how the plot moves, but we don’t know if they’re playing Sylvester Stallone, Millie Bobby Brown, or Brown. An old woman named Helen.
As we explore our creative potential in this new paradigm, and as machine learning and AI become more powerful and sophisticated, it looks like we’ll start to see these technologies not only generate linear content, but interactive content as well. Like games flappy bird It only consisted of a few hundred lines of code, and when it was released in 2013, the speed at which people created clones of the game was astounding. But computers are fast. It won’t be long before machine learning can be used to generate these kinds of simple games, and soon platform owners will release proprietary generators that allow users to create games on demand: “A side-scroller in which I’m an ostrich in a tuxedo trying to escape a robot uprising.” This may sound strange, and it may be pointless, but I wouldn’t be surprised if it showed up. tomorrow. And once, it will be better.
It has the potential to generate simple games that are not far off from this: “An open-world fantasy RPG in a steampunk version of First Napoleon France.” Of course, these are very broad parameters – more extensive than the remake of the mentioned script – but nothing prevents you from checking the parameters while playing; “Make the growth look more like that Skyrim” or “Make the bosses like you more Elden ring” and AI will learn about you and your preferences over time.
Of course, if this is content that can be corrupted, what’s the difference between an animated movie and a game? Between Jordan Pelly A demolition manCan’t I just say “let me play now” and immediately take control of Millie Bobby Brown’s Officer Spartan? While I can’t tell you how to build this thing, I can definitely tell you that it comes across as a script, not as a linear sequence of images rendered frame by frame. You can definitely play.
Given the tidal wave of content that seems poised to destroy our culture, it’s easy to fear a time when artists will be replaced by machines, but it’s not easy. There are two kinds of artists: creators who make nouns beautiful, and performers who make verbs beautiful (most artists are at least a little bit of both). In the last century, with the rise of broadcast media culture, the industrialization of entertainment, and the central control of the economy of created artifacts, creators have benefited more than performers, but this is historically ambiguous. For the first 10,000-ish years of human culture, the balance has been tipped and the correction is underway after Napster blew the music industry below the waterline and the concept of albums meant to be listened to linearly on plastic discs no longer made sense. .
In the future – as early as 2032 – the process of creating beautiful digital names will be completely automated. This does not eliminate our desire or obligation to continue operating digital names (even if it challenges our ability to do so profitably). Nor does it affect our desire to play… to play. No computer needs to beat a game of chess or beat a human, but that doesn’t stop us from wanting to play, play better, or play beautifully. As humans, we find beauty and wisdom in what other people do, and our fascination with pushing the boundaries of who we are and what we are capable of is inevitable.