People consider creativity to be inherently human. However, artificial intelligence (AI) has reached the stage where it could be creative as well.
A recent competition attracted anger from artists after it awarded a prize to an artwork created by an AI model referred to as Midjourney. And such software is now freely available because of the discharge of an analogous model called Stable Diffusion, which is probably the most efficient of its kind to this point.
Unions of creative practitioners equivalent to Stop AI Stealing the Show have for a while been raising concerns in regards to the use of AI in creative fields. But could AI actually replace human artists?
Author provided
These latest AI models can produce infinite possibilities. Each image of the robots shown above are unique, yet are generated by Stable Diffusion from similar user requests.
There are two ways to make use of these AI artists: write a brief text prompt, or provide a picture alongside the prompt to offer more guidance. From a 14-word prompt, I used to be capable of generate several logo ideas for a made-up company that delivers fruit. In just below 20 minutes. On my mid-range laptop.
As you’ll be able to see from the outcomes above, Stable Diffusion struggles to create art involving words. And among the fruit are a bit funky.
Yet there is no such thing as a way I could have produced anything remotely like this without using AI or employing the assistance of a graphic designer. I couldn’t have created the robot pictures myself either.
The potential of this technology hasn’t gone unnoticed – the startup chargeable for Stable Diffusion, Stability AI, is targeting a US$1 billion (£900 million) investment evaluation. But these AI models are starting to have an effect in the actual world, as seen with the prize-winning Midjourney picture. Indeed where AI really excels is producing pieces of tremendous art that mix different elements and styles.
Yet while AI may do many of the legwork for you, using these models still requires skill. Sometimes a prompt doesn’t generate quite the image that you just wanted. Or the AI might be used alongside other tools, only making up a small part of a bigger pipeline.
And generating tremendous art is different to producing digital designs. Stable Diffusion is healthier at drawing landscapes than logos.
Why Stable Diffusion is a game changer
AI models are typically trained to create art using a dataset containing a staggering 5.85 billion images. This vast amount of knowledge is required so the AI can find out about image content and artistic concepts. And it takes a really very long time to process.
For Stable Diffusion, it took 150,000 hours (just over 17 years) of processor time. However, this may reduced to lower than a month of real time by training in parallel on large compute clusters (collections of powerful computers that act a single device).
Stability AI also provides an internet tool called DreamStudio that lets you use its AI model at a price of around US$0.01 per image. In comparison, to make use of competitor OpenAI’s art model, DALL·E 2, the price is over ten times that.
Both methods use the identical underlying approach, referred to as a diffusion model computer program, which learns to create latest images by taking a look at plenty of existing images. However, Stable Diffusion has a lower computational cost, meaning it requires less time to coach, and uses less energy.
Plus, you’ll be able to’t actually download and run OpenAI’s model yourself, only interact with it via an internet site. Stable Diffusion, meanwhile, is an open-source project that anyone can mess around with. So it enjoys the advantage of rapid development by the net coding community, equivalent to improvements to the models, user-guides, integration with other tools. This has already been happening within the weeks after Stable Diffusion’s release in August 2022.
The way forward for art?
While vast improvements have been made within the last five years, there are still things that AI art models struggle with. Words of their artworks are recognisable but often gibberish. Similarly, AI struggles to render human hands.
There’s also the plain constraint that these models can only produce digital art. They can’t work with oils or pastels like people can. In the way in which that vinyl has made a comeback, technology may initially create a swing towards a brand new form, but over time people all the time appear to circle back to the unique form with the highest quality.
Ultimately, as previous research has found, AI models of their current form usually tend to act as latest tools for artists than as digital replacements for creative humans. For example, the AI could generate a variety of images to function a place to begin, which may then be chosen from and improved by a human artist.
This combines the strengths of AI art models (rapid iteration and creation of images) with the strengths of human artists (a vision for the piece of art and overcoming the issues with AI models). This is particularly true within the case of commissioned art when a particular output is required. AI by itself is unlikely to provide what you wish.
However, there continues to be a danger for creatives. Digital artists who select not to make use of AI could also be left behind, unable to maintain up with the rapid iteration and lower costs of AI-enhanced artists.
This article was originally published at theconversation.com
