There has recently been a considerable influx in Internet content generated by artificial intelligence.
The software tools used to generate this content range dramatically. Some were formerly used in research projects at various big universities and tech companies, often requiring very powerful computers.
Others were distilled down to smartphone applications usable anywhere, thanks to cloud computing technology. Stable Diffusion, Dall-E and MidJourney are just some of the tools available today.
In the last few months alone, we’ve seen some very basic images being generated from text phrases advancing to the point now where the same text phrase can create short video clips and even write stories— with little to no input from the users other than the starting text.
The systems in use behind the scenes can now generate incredible scenes and artwork with photorealistic-looking imagery, to the extent
Sponsored Ads
where the users are even able to specify what style is being used—and even the type of camera and lenses to be used—in the creation of the artificial images. All without having to even own those real-world devices.
Other recent developments are the ability to upload your own images and have these tools reimagine them in a different style or enable you to erase part of an image to have the AI tools replace that erased part with something new.
Small cropped photos can be uploaded, and the tools will expand the canvas to show you what it thinks was outside the frame when the photo was taken. It truly is unbelievable
Musicians are able to train these systems on their voices and generate entirely new songs without having to actually sing into a microphone. Artist Holly Hearndon created an AI version of her voice that could perform a cover of Dolly Parton’s Jolene on its own, as the AI artist Holly+.
Creators of these tools are also starting to enable developers and companies to integrate AI into all kinds of different applications. Stock photo sites can now generate imagery from your search terms instead of relying on a back catalogue of existing photography.
Microsoft is integrating tools that enable you to add photos and artwork to your PowerPoint presentations without having to hire an illustrator or photographer.
Sponsored Ads
“Imagine being able to use Netflix to search for a movie starring your favourite actors, in your favourite genre, that doesn’t actually exist, but the AI can generate it for you in real time?”
So where is this all going?
Things are moving at such a rapid pace that it can be hard to speculate where things will be in the next few months—but it seems reasonable to assume that within a few years,
these tools will have advanced to the point where you’ll just use your voice to request something from the AI and it will be delivered directly to your smartphone. Imagine being able to use Netflix to search for a movie starring your favourite actors, in your favourite genre, that doesn’t actually exist, but the AI can generate it for you in real time? We’re not that far off. And with the rapid pace of improvement in these tools, that reality is coming sooner than you might think.
