A British physicist who once touched on the subject of artificial intelligence Stephen Hawking “Success in creating advanced artificial intelligence will be the greatest event in human history. Unfortunately, however, it may also be the last,” he continued, adding that the greatest danger to the survival of our species on planet Earth is not in global warming or a nuclear holocaust, just as Schwarzenegger would invent an artificial intelligence advanced enough to rebel against our control, as in the Terminator movies.
These days, after all PCs are still in crisis when you try to set up a new printer, and Boston Dynamics robots are notorious for their devastating fall for their practical applications, the possibility of being wiped out by a computer may seem particularly remote. Truth be told, the technology surrounding artificial intelligence has only made an extraordinary evolutionary leap over the past year and a half, and some organizations are investigating if the neural networks of natural language reprocessing are no longer indistinguishable from a human interlocutor. like OpenAI They went even further and developed artificial intelligence that can assign meaning to pictures as well as words.
Starting from a simple textual description, we are talking about DALL-E 2, an artificial intelligence that has become popular in recent months with its extraordinary ability to process an image, whether realistic or purely artistic. An anime dedicated to Sonic’s financial trading? Or did Mario deal with the compilation of taxes? No matter how absurd or probable the input we feed it, plexus It metabolizes it to create a graphic representation of what we want out of nothing.
In this article we will discover Possible applications of DALL-E 2 especially in the field of video gamesWe will try to understand what the future role of next generation artificial intelligences with almost unlimited potential could be in the development of these fun experiences for our grandchildren to play. Provided, of course, that you are not yet a slave to the machines.
Text to image in a fraction of a second
In fact, besides the DALL-E 2, there are several AIs that offer the ability to: get image from a text prompteach has its own strengths and weaknesses. Developed by OpenAI, it is currently in beta and can only be accessed by signing up for a very long waiting list, but among all, it is the most effective at rendering realistic images inspired by real photographs and has no equivalent in the representation of faces. .
Then there is travelInstead, it’s already available to anyone with a Discord account: you enter the AI channel and you can start trying right away, even if the trials are limited, and you’ll be prompted to subscribe once you’ve exhausted them. Midjourney is very skilled at creating detailings with a strong artistic impact, has a preference for theatricality and will therefore be perfect for helping developers create gorgeous concept art that will inspire them.
But what makes AIs like DALL-E 2 and Midjourney extraordinary is their ability to picture something that doesn’t exist in the world. creative vision an idea, a concept, just as a human artist would. There is nothing exceptional about processing a photo of a dog with a top hat and a cigar, but this is when AI is asked to derive a representation of hell, heaven, the limits of the universe, or the center of a black hole. , which the situation changes drastically. In this case, the neural network is absorbing the text to give it its own text. free translation and this is as surprising as it is vaguely disturbing.
Apps
Clearly, DALL-E 2 and other neural networks like OpenAI’s are nearly endless. apps, especially in the field of design and graphics, but also in the field of marketing and all industries where images are used to convey a message. Of course, it is not mandatory to use them for purely professional purposes only, for example we look forward to using them to create illustrations of the characters of a Dungeons & Dragons campaign or to decorate the desktop with wallpapers that are unique in their kind.
when it comes video gamesThere are just as many applications of neural networks and artificial intelligence, and no, they do not suppress the creativity of artists, as many of you will immediately point out in the comments section at the bottom of the article. For example, consider building a game world and the amount of it. repetitive textures and other assets in an open world. An AI can process the 2D surfaces of streets and buildings, or the details of roofs, or even the facades of stores, in a matter of minutes, making each item different from the previous one, without an artist needing hundreds of hours. person. Starfield with over 1000 celestial bodies with unique planetary surfaces? It’s a productive nightmare for a software house, but a walk-through for AI.
DALL-E 2 also creative process that precedes actual development, providing ideas for character design, or providing concept art to the team involved in building levels. By supporting developers in their various needs from time to time, an AI can significantly streamline the production processes surrounding the creation of a video game, make boredom less frequent, and provide a better work experience to all of the studio’s employees without changing it. none.
what awaits us
Thinking about the future of AI’s application in video game development is pretty straightforward, because, as Hawking himself said in his talk about artificial intelligences mentioned at the outset, there is no real limit that can hinder its evolution. If today they can already interpret a text to turn it into an image, there is no reason to doubt that they will continue to elaborate a bit more in the near future. Complete 3D models, they create all parts of the game map on their own, starting with the instructions from the developers. Something similar to this already exists today regarding the composition of virtual forests, but after a while we may see the further spread of these systems and as a result it will become increasingly easy to develop larger experiences despite their size.
Looking further, we couldn’t escape the imagination by collecting Hawking’s indicators and sketching out a distant but not-so-distant future where our grandchildren could sit at the PC and ask AI to do it. them in place nonexistent video gameMixing typical elements of a la Miyazaki experience with the haunting aesthetic of Night City, it takes the very young Arthur Morgan as the hero, wearing teardrop glasses and a robotic arm.
The British physicist was convinced that the problem was not with him, but only with when. And we really like to believe it.
What do you think about this topic? What do you want the DALL-E 2 to achieve? And what do you think will be the horizons of artificial intelligence in the next 50 years?
Source: Multiplayer
