AI can now create video games as Google teaches it to make Doom clones-Kenneth Andersen-Entertainment – Metro

A sign of what’s to come? (Xbox)

A significant step forward has been made in video game development, as Google replicates Doom using artificial intelligence.

Artificial intelligence is becoming more and more widespread in society, and the video games industry is no different, with many voice actors worried for their careers after their union agreed a partnership with an AI voice technology company.

This was exacerbated by a recent report that suggested developers are using generative AI to make games, with artists and graphic designers being most affected (i.e. their losing their jobs) by the growing trend.

Now, Google has taken things to the next level, as it has managed to get an AI to make a version of the classic shooter Doom, even if it only runs at 20fps.

A Google research team has managed to create a neural network – a method that teaches computers to process data in a way that is inspired by the human brain – that can create real-time gameplay of Doom, using its engine GameNGen.

The process doesn’t use a traditional game engine full of code, but simulates the entire game environment using an AI-driven generative diffusion model.

According to the paper released by the team, GameNGen is ‘the first game engine powered entirely by a neural model that enables real-time interaction with a complex environment over long trajectories at high quality’.

The AI (GameNGen) was trained in two phases to replicate Doom: to learn to play the game and to use a diffusion model to create the next frame, based on the sequence of previous frames and actions.

Shlomi Fruchter, of Google DeepMind, uploaded a video of people playing the AI version of Doom on YouTube, as seen below.

According to the paper, the AI version of Doom has a ‘slightly better than random chance’ of people being able to tell the difference from the actual game.

The AI still needs someone to play the game, so that it can learn and replicate it, which is how other forms of generative AI works, such as ChatGPT and tools that can generate images from what you feed it.

AI has been used in video games since the very start, but as the technology improves, many within the games industry are worried that it comes at the cost of jobs and quality.

A recent example of how sensitive the subject is, was when Palworld was accused of training AI to create Pokémon-like creatures, which resulted in staff at developer Pocketpair receiving death threats online.

The future of AI remains uncertain (Wikipedia)

Email gamecentral@metro.co.uk, leave a comment below, follow us on Twitter, and sign-up to our newsletter.

To submit Inbox letters and Reader’s Features more easily, without the need to send an email, just use our Submit Stuff page here.

For more stories like this, check our Gaming page.

MORE : EA Sports College Football 25 is the future of AI video game development

MORE : Video game strike warnings and Call Of Duty controversy as AI claims more jobs

MORE : A new petition needs 1,000,000 signatures to stop publishers killing games

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.Entertainment – MetroRead More