Artificial intelligence (AI) has revolutionised the way humans interact, work, and consume content. The use of AI has increased in pretty much all aspects of life; not just in industries, but also at home in the form of AI-powered digital assistants and home appliances.
And AI now has a new application--in the creation of art. With non-fungible tokens (NFTs) taking the world of art and music by storm, it is hardly a surprise that AI could become the next big leap in the space. AI will reportedly now be used to generate art-like images in the form of NFTs--a type of non-interchangeable digital ledgers that allow creators to sell their art in the digital format.
The NFT market gained much popularity last year, with the market capitalisation crossing $41 billion in 2021. NFTs have helped creators gain freedom from traditional financing solutions. With the adoption of AI technology, NFTs allow artists to create their pieces by just entering keywords into the system.
But how does AI create artwork?
The AI generates these art-like NFTs by using Generative Adversarial Networks or GANs. These smart programs are heavily programmed but more importantly, do not require users to have any coding skills.
The concept of using AI to create art is relatively new but has shown massive potential with multiple startups entering the business since 2018. Art AI, for example, is one of the largest AI art galleries. The company recently launched Eponym, an AI software that converts text to art-like images and helps the artist to sell their pieces directly on NFT marketplace OpenSea. The art pieces produced using Eponym have been launched on the ethereum blockchain. Eponym is different from traditional AI-based art as it allows a human touch and is not totally based on AI algorithms.
According to Art AI co-founder Eyal Fisher, the majority of Eponym users are creating NFTs for the first time. NFT collectors and buyers are making their own creations--a gamechanger in the space since most people part of the Eponym community are not professional artists, he told CoinTelegraph.
“Eponym is a collaborative NFT project. Users access it by coming to the website and typing any phrase or word into a text box. The AI then generates artwork based on the text entered.” Fisher said, adding that each text prompt can only be generated once.
Also read: Storyboard18 | 2022 - Definitely, Maybe: Year of creators, NFTs, DEI and sustainable brands
Another AI-based program called MetaScapes combines photographs to generate a new image. It runs on two learning models, and hence, the output keeps getting better with time. The MetaScape is made up of a generator that provides an output and a discriminator that reviews the workings of the generator and instructs it on how to improve the output. The users of the MetaScape would need to input photographs that are labelled in relation to the input imagery. The AI program then analyses the photographs and their labels to produce new images that are a culmination of the submitted images.
Ryan Newburn, one of the three photographers behind Metascape, told Cointelegraph that the project consists of 3,333 rare AI-generated NFTs based on photographs taken across the world.
This newfound freedom to create digital assets (NFTs) without the know-how of complex computing languages has motivated many people to create NFTs for the first time. Since NFTs can be sold on online marketplaces, creating these digital tokens helps people earn money from the comfort of their homes.
The use of AI technology to create art is only going to increase. Experts predict that the integration of AI in the creation of NFTs would reach a level where users can chat with the program to create art. The increasing popularity of the metaverse or the virtual universe, is also likely to propel AI-based NFTs.
Eponym’s next project will use AI to create interactive 3D avatars for users through their portraits. “Our idea is to use AI that will allow for avatars to take different shapes that are compatible in metaverse environments like Sandbox. In February, the company will be introducing additional algorithms that will allow users to generate personalized avatars.” Fisher told CoinTelegraph.
However, since AI technology has only recently found application to the generation of art, there is still a long way to go until we can entirely rely on the program to create meaningful art. The process of generating art pieces using AI is currently being run on the trial-and-error principle where all the program outputs are reviewed and re-run to generate better results.
First Published: IST