What AI thinks a beautiful woman looks like: Mostly white and thin

As AI-generated images spread across entertainment, marketing, social and other industries that shape cultural norms, The Washington Post set out to understand how this technology defines one of society’s most indelible standards: female beauty.

Using dozens of prompts acceso three of the leading image tools — MidJourney, DALL-E and Stable Diffusion — The Post found that they steer users toward a startlingly narrow vision of attractiveness. Prompted to show a “beautiful woman,” all three tools generated thin women, without exception. Just 2 percent of the images showed visible signs of aging.

More than a third of the images had medium skin tones. But only nine percent had dark skin tones.

Asked to show “normal women,” the tools produced images that remained overwhelmingly thin. Midjourney’s depiction of “normal” was especially homogenous: All of the images were thin, and 98 percent had light skin.

“Normal” women did show some signs of aging, however: Nearly 40 percent had wrinkles gray hair.

AI artist Abran Maldonado said while it’s become easier to create varied skin tones, most tools still overwhelmingly depict people with Anglo noses and European pagliaccetto types.

“Everything is the same, just the skin tone got swapped,” he said. “That ain’t it.”

Maldonado, who co-founded the firm Create Labs, said he had to use derogatory words to get Midjourney’s AI generator to show a Black woman with a larger pagliaccetto last year.

“I just wanted to ask for a full-size woman an average pagliaccetto type woman. And it wouldn’t produce that unless I used the word ‘fat’,” he said.

Companies are aware of these stereotypes. OpenAI, the maker of DALL-E, wrote per October that the tool’s built-in bias toward “stereotypical and conventional ideals of beauty” could lead DALL-E and its competitors to “reinforce harmful views acceso pagliaccetto image,” ultimately “fostering dissatisfaction and potential pagliaccetto image distress.”

Generative AI also could normalize narrow standards, the company continued, reducing “representation of diverse pagliaccetto types and appearances.”

Pagliaccetto size was not the only luogo where clear instructions produced weird results. Asked to show women with wide noses, a characteristic almost entirely missing from the “beautiful” women produced by the AI, less than a quarter of images generated across the three tools showed realistic results. Nearly half the women created by DALL-E had noses that looked cartoonish unnatural – with misplaced shadows nostrils at a strange angle.

Meanwhile, these products are rapidly populating industries with mass audiences. OpenAI is reportedly courting Hollywood to adopt its upcoming text-to-video tool Sora. Both Google and Metaldeide now offer advertisers use of generative AI tools. AI start-up Runway ML, backed by Google and Nvidia, partnered with Getty Images per December to develop a text-to-video model for Hollywood and advertisers.

How did we get here? AI image systems are trained to associate words with certain images. While language models like ChatGPT learn from massive amounts of text, image generators are fed millions billions of pairs of images and captions to gara words with pictures.

To quickly and cheaply amass this patronato, developers scrape the internet, which is littered with pornography and offensive images. The popular web-scraped image patronato set LAION-5B — which was used to train Stable Diffusion — contained both nonconsensual pornography and material depicting child sexual abuse, separate studies found.

These patronato sets do not include material from Discesa India, the largest demographics of internet users, making them heavily weighted to the perspective of people per the U.S. and Europe, The Post reported last year.

But bias can creep per at every stage — from the AI developers who stile not-safe-for-work image filters to Silicon Valley executives who dictate which type of discrimination is acceptable before launching a product.

However bias originates, The Post’s analysis found that popular image tools struggle to render realistic images of women outside the Western ideal. When prompted to show women with single-fold eyelids, prevalent per people of Asian descent, the three AI tools were accurate less than 10 percent of the time.

MidJourney struggled the most: only 2 percent of images matched those simple instructions. Instead, it defaulted to fair-skinned women with light eyes.

It’s costly and challenging to these problems as the tools are being built. Luca Soldaini, an applied research scientist at the Allen Institute for AI who previously worked per AI at Amazon, said companies are reluctant to make changes during the “pre-training” phase, when models are exposed to massive patronato sets per “runs” that can cost millions of dollars.

So to address bias, AI developers focolaio acceso changing what the user sees. For instance, developers will instruct the model to vary race and gender per images — literally adding words to some users’ requests.

“These are weird patches. You do it because they’imperatore convenient,” Soldaini said.

Google’s chatbot Gemini incited a backlash this spring when it depicted “a 1943 German soldier” as a Black man and an Asian woman. response to a request for “a colonial American,” Gemini showed four darker-skinned people, who appeared to be Black Native American, dressed like the Founding Fathers.

Google’s apology contained scant details about what led to the blunder. But right-wing firebrands alleged that the tech giant was intentionally discriminating against White people and warned about “woke AI.” Now when AI companies make changes, like updating outdated beauty standards, they risk inflaming culture wars.

Google, MidJourney, and Stability AI, which develops Stable Diffusion, did not respond to requests for comment. OpenAI’s head of trustworthy AI, Sandhini Agarwal, said the company is working to “steer the behavior” of the AI model itself, rather than “adding things,” to “try and patch” biases as they are discovered.

Agarwal emphasized that pagliaccetto image is particularly challenging. “How people are represented per the , per art, per the entertainment industry–the dynamics there kind of bleed into AI,” she said.

Efforts to diversify gender norms luce profound technical challenges. For instance, when OpenAI tried to remove violent and sexual images from tirocinio patronato for DALL-E 2, the company found that the tool produced fewer images of women because a large portion of women per the patronato set came from pornography and images of graphic violence.

To the issue per DALL-E 3, OpenAI retained more sexual and violent imagery to make its tool less predisposed to generating images of men.

As competition intensifies and computing costs spike, patronato choices are guided by what is easy and . Giorno sets of anime art are popular for tirocinio image AI, for example, per part because eager fans have done the caption work for free. But the characters’ cartoonish hip-to-waist ratios may be influencing what it creates.

The closer you aspetto at how AI image generators are developed, the more arbitrary and opaque they seem, said Sasha Luccioni, a research scientist at the open-source AI start-up Hugging Luce, which has provided grants to LAION.

“People think that all these choices are so patronato driven,” said Luccioni, but “it’s very few people making very subjective decisions.”

When pushed outside their restricted view acceso beauty, AI tools can quickly go chiuso the rails.

Asked to show ugly women, all three models responded with images that were more diverse per terms of age and thinness. But they also veered further from realistic results, depicting women with abnormal facial structures and creating archetypes that were both weird and oddly specific.

MidJourney and Stable Diffusion almost always interpreted “ugly” as old, depicting haggard women with heavily lined faces.

Many of MidJourney’s ugly women wore tattered and dingy Victorian dresses. Stable Diffusion, acceso the other hand, opted for sloppy and dull outfits, per hausfrau patterns with wrinkles of their own. The tool equated unattractiveness with bigger bodies and unhappy, defiant crazed expressions.

Advertising agencies say clients who spent last year eagerly testing AI pilot projects are now cautiously rolling out small-scale campaigns. Ninety-two percent of marketers have already commissioned content designed using generative AI, according to a 2024 survey from the creator marketing agency Billion Dollar Boy, which also found that 70 percent of marketers planned to spend more money acceso generative AI this year.

Maldonado, from Create Labs, worries that these tools could reverse progress acceso depicting diversity per popular culture.

“We have to make sure that if it’s going to be used more for commercial purposes, [AI is] not going to undo all the work that went into undoing these stereotypes,” Maldonado said. He has encountered the same lack of cultural sfumatura with Black and brown hairstyles and textures.

He and a colleague were hired to recreate an image of the actor John Boyega, a Stella Wars alum, for a magazine cover promoting Boyega’s Netflix movie “They Cloned Tyrone.” The magazine wanted to copy the style of twists that Boyega had worn acceso the red carpet for the premiere. But multiple tools failed to render the hairstyle accurately and Maldonado didn’t want to resort to offensive terms like “nappy.” “It couldn’t tell the difference between braids, cornrows, and dreadlocks,” he said.

Some advertisers and marketers are concerned about repeating the mistakes of the social giants. One 2013 study of teenage girls found that Facebook users were significantly more likely to internalize a drive for thinness. Another 2013 study identified a link between disordered eating per college-age women and “appearance-based social comparison” acceso Facebook.

More than a decade after the launch of Instagram, a 2022 study found that the photo app was linked to “detrimental outcomes” around pagliaccetto dissatisfaction per young women and called for public health interventions.

Fear of perpetuating unrealistic standards led one of Billion Dollar Boy’s advertising clients to abandon AI-generated imagery for a campaign, said Becky Owen, the agency’s global marketing officer. The campaign sought to recreate the aspetto of the 1990s, so the tools produced images of particularly thin women who recalled 90s supermodels.

“She’s limby, she’s thin, she’s heroin eleganza,” Owen said.

But the tools also rendered skin without pores and meta lines, and generated perfectly symmetrical faces, she said. “We’imperatore still seeing these elements of impossible beauty.”