If you’ve been around social media lately, you might have noticed some of your friends posting otherworldly digitally animated portraits of themselves.
Even celebrities like luck the rapper and Megan Fox jumped on the bandwagon.
The portraits look like they were made by digital artists, and in a way, they kind of were. But no individual artist is paid for it, or even credited.
This is because these works were not created by a human. They were created by a machine.
Prisma Labs’ Lensa AI app uses artificial intelligence to turn your selfies into personalized portraits, empowering users to be who they want to be.
But while it has become a runaway hit on social media, the app has drawn the ire of digital artists, who claim the works it generates are based on stolen art.
New to the story? Here’s what you need to know.
How does Lensa work for users?
Creating these digital portraits using Lensa AI is simple and relatively inexpensive, two facts that have no doubt contributed to the app’s meteoric rise in recent weeks.
According to Statista, Lensa AI has been downloaded over 5 million times since November, when the “magic avatar” feature was introduced.
A subscription normally costs around 100 euros per year, but users can benefit from a 7-day free trial. Access to the “magical avatar” function, which generates 50 fantastic portraits, costs an additional 4 euros each.
To access their portraits, users need to upload 10-20 selfies of themselves to the app.
Lensa urges them to choose a variety of different angles and facial expressions for best results, and warns them that there may be “inaccuracies or flaws” in the output images.
These flaws sometimes show up as extra arms and legs, or heads that turn at unnatural angles.
After the app has had time to run a user’s photos through its AI model (about 10 minutes), it spits out a variety of avatars. Users can then save them to their devices and share them freely on all social media platforms.
Regarding data protection, Prisma Labs wrote in a long FAQ that user photos are deleted from its servers immediately after avatars are generated.
The avatars themselves are stored “for as long as it takes to provide the service to our users” and Prisma Labs claims that all avatars belong to the user who created them.
Why are artists worried about Lensa?
As millions of users around the world have fallen in love with their magical avatars, a wave of anxiety has begun to sweep through artist communities online.
Not only did these AI-generated portraits remove commissioning opportunities for digital artists, but some of these artists’ works were used to train the AI model that generated them… often without their permission.
On Instagram, digital artist Meg Rae posted a warning that was widely shared on the app:
“Do not use the Lensa app’s ‘Magic Avatar’ generator,” she wrote. “It uses Stable Diffusion, an AI art model, to sample works from artists who have never consented to their work being used. It is art theft”.
It is true that Lensa uses a copy of the open source Stable Diffusion neural network model to train its AI. The model draws from a pool of billions of images from all corners of the internet, which are compiled into a dataset called LAION-5B.
Stable Diffusion then uses these images to learn techniques that it applies to generate new works, which Lensa says “are not replicas of any particular artist’s work”.
The copyright law regarding these datasets is obscure. of LAION website says that because the datasets only contain image URLs, they serve as an index on the Internet, which does not violate copyright law.
The CEO of Stability AI, creator of Stable Diffusion, also defended the AI model and its machine learning process.
Emad Mostaque told The Verge that the scraping of public images from the Internet – even copyrighted ones – was legal in the US and UK and that the open source nature of Stable Diffusion meant that any copyright infringement was the responsibility of the end user.
But while AI art can remove legal hurdles, the ethics surrounding these image generators are far from straightforward.
What can artists do to regain control?
The virality of Lensa AI’s magical avatars has sparked a conversation among artists about how they can take back control of their work in the age of machine learning.
One of the tools at their disposal is the search engine haveibeentrained.com, developed by Berlin artists Mat Dryhurst and Holly Herndon.
It allows artists to see if their work is included in the LAION datasets used to train the AI, and then opt out if they wish.
Dryhurst told Euronews Culture via email that Stability AI and LAION have agreed to honor any unsubscribe requests made by artists.
European Union Data Protection Law (GDPR) also allows EU citizens to contact LAION directly to request removal of any identifying information, such as their name or likeness, from all of data.
But Dryhurst said beyond the technical opt-out procedure, there needs to be a broader conversation about consent and collaboration in the art world.
“A new consent protocol, which is our ultimate interest, is easier to imagine executed technically than it is socially, but the social dimension is probably the most important,” Dryhurst told Euronews Culture. “We hope this new suite of tools will lead to more attribution and sharing of benefits between artists, and a greater discursive shift towards recognizing the interdependence of creative worlds.”
Is AI inherently anti-artist?
“I’m not against artificial intelligence, I want it to be clear,” said Spanish artist Amaya Díaz. “If it’s a tool that we can use, and if people learn to value what we put into our work, I think that’s fine.”
Díaz, a Spanish cartoonist and illustrator who goes through Stelladiasaid Lensa’s frenzy inspired them to search for their works on haveibeentrained.com.
They found some of their very old artwork in the dataset they posted on the DeviantArt platform when they were students.
“I wasn’t really surprised,” Díaz told Euronews Culture. “As artists, we are very used to our work being used without consent. I think the reason why it becomes so important is because we are very exhausted.
Being an artist is hard, often thankless work, Díaz said. It takes years to hone their skills, getting paid fairly and on time is a constant struggle and there’s a lot of self-promotion involved which goes against the introverted nature of many artists.
“When you see your work being used for something that will maybe put you out of the market, it’s really exhausting,” they said. “You just feel helpless and you just want to go to bed and stop trying.”
Dryhurst said he thinks the recent uproar around AI-generated art is likely due to growing frustration with how society has devalued art over the past century.
“I think we’re still in the dress rehearsal period (of AI art), so there’s no reason to panic just yet,” he said. “I think it’s important for artists to assert control over their data, and I think this will create the preconditions for all kinds of new and exciting configurations to emerge.”
For Díaz, the hope is that as artificial intelligence evolves, so do human mindsets.
“Artificial intelligence is just a very empty mix of things that already exist,” they said. “If you look closely, you will see that there is no conscience behind it. I really hope people can understand what we (artists) are doing when we create something […] and I hope people will look forward to human art”.