Generative artificial intelligence (AI) has the potential to revolutionise fashion design. By recognising patterns in data and generating new text and images, AI models powered by deep learning algorithms can help fashion designers develop new catalogues, expanding creativity and helping bring products to market faster, as per a study by Pusan National University.

Large language models (LLMs) like ChatGPT and AI image generators like DALL-E have shown promising results across industries and popularised the use of AI. In fashion, LLMs can help designers and non-experts understand past styles and predict future fashion trends. These insights can then generate prompts for AI image generators to produce real fashion collections. As such, it is increasingly important to understand how AI can be effectively integrated into fashion.

In a recent study, professor Yoon Kyung Lee and master’s student Chaehi Ryu, from the Department of Clothing and Textiles at Pusan National University, South Korea, explored how generative AI can contribute to visualising seasonal fashion trends. “To use AI effectively in fashion, we must understand the characteristics of generative AI models and make informed judgements of where they can be applied,” explained. Lee. “In this study, we studied how effective prompt engineering can be used to generate realistic fashion collection images through AI.”

A study by Pusan National University shows that generative AI, using tools like ChatGPT and DALL-E 3, can help visualise and predict fashion trends.
By analysing past data and crafting precise prompts, AI generated realistic Fall/Winter 2024 men’s fashion images.
While effective, limitations remain, highlighting the need for expert input.

Using ChatGPT-3.5 and ChatGPT-4, the researchers first analysed men’s fashion trends, based on historical data up to September 2021. From this, they used ChatGPT to predict men’s fashion trends for Fall/Winter 2024. Design elements from these predictions were classified as ‘initial codes’. In addition, design elements from Vogue’s 2024 Fall/Winter Men’s Fashion Trend data were used as ‘modified codes’, and those from literature as ‘codes from literature’. These were then regrouped into six final codes: trends, silhouette elements, materials, key items, garment details, and embellishments.

Using these codes, they created 35 prompts for DALL-E 3, each describing a unique outfit. The prompts followed a consistent template featuring a male model walking down a runway at a 2024 Fall/Winter fashion show. The template allowed customisation of event details, including aspect ratios, events, camera angles, model appearance and height, runway design, background, and audience details, and moods. Each prompt was run three times, generating a total of 105 images.

DALL-E 3 was able to perfectly implement the prompts 67.6 per cent of the time. Prompts with adjectives demonstrated a high implementation rate. Some generated images closely resembled actual 2024 Fall/Winter Men’s fashion collections. However, there were errors—most leaned toward ready-to-wear fashion, and DALL-E struggled with trend elements like gender fluidity. Trend keywords alone were insufficient to generate accurate results, indicating a need for further learning.

“Our results show that expertly worded prompts are necessary for accurate fashion design implementation of generative AI, highlighting the important role of fashion experts,” added Lee. “With further learning and improvements, generative AI models like DALL-E 3 will help fashion designers create entire fashion collections more efficiently, while supporting their creativity, and also help non-experts understand fashion trends.”

The study shows that generative AI can be a powerful tool not just for professionals but also for the general public, making it easier than ever to explore, predict, and style the upcoming season’s fashion with confidence.

Fibre2Fashion News Desk (RR)

In another vote, advisers recommended adding language on the shot’s risks to the vaccine’s information sheet, which is already required by law.

The committee’s focus on Covid-19 vaccines reflects Kennedy’s long-held suspicion of them. Since taking office in February, Kennedy has canceled a half-billion dollars in mRNA vaccine research and separately ended a major contract with Moderna, one of the Covid vaccine manufactures, for work on a pandemic bird flu vaccine.

During Friday’s meeting, CDC scientists presented extensive data on the safety and efficacy of the Covid vaccines. They also explained in detail how the agency tracks Covid hospitalizations and said the agency has a “rigorous and standardized process” to determine whether hospitalizations are classified as being due to Covid-19.

During the discussion portion of the meeting, committee members made several unfounded claims. Robert Malone, a former mRNA researcher who has spread vaccine misinformation, questioned whether there is actually evidence of disease protection from the Covid shots. “Are there any well-defined, characterized correlates of protection for Covid, yes or no?” he demanded.

Cody Meissner, a pediatrician at Dartmouth College, responded that there is “a reasonable measurement of neutralizing or binding antibodies that correlate with protection against symptomatic infection in the first few months” after vaccination.

At one point, Hilary Blackburn, a pharmacist on the committee, questioned whether the Covid vaccine could be connected to her mother’s lung cancer diagnosis, which occurred two years after receiving a Covid vaccine. She said she is aware of four other individuals in her small hometown diagnosed with the same kind of cancer. “Is it related to the vaccine?” she asked.

In a tense exchange about potential birth defects associated with the Covid vaccines, some ACIP members pressed manufacturer Pfizer about eight birth defects that occurred in a group of pregnant women who received the company’s vaccine and two birth defects that occurred in an unvaccinated group. Alejandra Gurtman, who heads vaccine clinical research and development at Pfizer, replied that those rates are comparable to rates of congenital abnormalities seen in the general population.

Carol Hayes, a liaison with the American College of Nurse-Midwives who was present during the meeting, clarified that most birth defects arise during the first trimester of pregnancy, and in the cited study, mothers received the vaccine at 12 to 24 weeks of pregnancy.

At Friday’s meeting, the committee also reversed a decision it made just a day before. On Thursday, advisers voted to no longer recommend the combined measles, mumps, rubella, and varicella (MMRV) vaccine to children under age 4. Yet puzzlingly, it voted to maintain coverage of that vaccine through the federal Vaccines for Children program, which provides free vaccines to low-income children and those without insurance. On Friday, they voted that the program should not, in fact, cover it.

On Friday, advisers also voted 11 to one in favor of tabling a decision on whether to delay the birth dose of the hepatitis B vaccine until one month of age. The committee had discussed that vaccine extensively on Thursday, though it’s unclear why the committee was asked to look into the potential change at all, as the hepatitis B vaccine has been given to newborns in the US since 1991.

Infants get the vaccine before leaving the hospital because the virus can be passed from an infected mother to the baby during birth. Hepatitis B is a serious liver infection that can lead to cirrhosis and cancer. The vaccine is highly effective at preventing infection in newborns.

Chari Cohen, president of the Hepatitis B Foundation, tells WIRED there is no scientific rationale for delaying the hepatitis B vaccine until one month after birth and she worries about an increase in hepatitis B infections if the panel eventually recommends delaying the immunization.

“We will likely see more babies and young children who become infected,” Cohen says. “From a public health infrastructure perspective, we are concerned that this risk-based approach will miss preventing infection to babies born to infected moms.”

Up to 16 percent of HBV-positive pregnant women don’t get tested for hepatitis B, so screening doesn’t capture all infected mothers.

“We do not understand the motivation or rationale for this debate,” Cohen says.

The United States and China may have agreed on a deal to prevent the social platform TikTok from being banned in the US—if you take US president Donald Trump’s word for it. After a long-awaited call between Trump and Chinese president Xi Jinping on Friday, Trump announced victory on Truth Social: “The call was a very good one, we will be speaking again by phone, appreciate the TikTok approval, and both look forward to meeting at APEC!”

As for any details on the agreement, good luck. Specifics around the shape and scope of the deal remain largely unclear as of Friday afternoon. More importantly, there’s been no official word from the Chinese government on whether it has agreed to the terms.

“China’s position on the TikTok issue is clear: The Chinese government respects the wishes of the company in question and would be happy to see productive commercial negotiations in keeping with market rules lead to a solution that complies with China’s laws and regulations and takes into account the interests of both sides,” says China’s official readout of the call, which was posted on the website of China’s Ministry of Foreign Affairs.

The deal being proposed by the Trump administration involves Oracle, Silver Lake, and Andreessen Horowitz leading a group of investors to take a roughly 80 percent stake in TikTok’s US operation, according to The Wall Street Journal. Oracle, which has worked closely with TikTok since 2020, would continue to store US user data on its domestic servers. The new, US-controlled entity would use licensed technology from ByteDance, TikTok’s Chinese parent company, to create a similar content recommendation algorithm to the one TikTok currently employs.

“Any details of the TikTok framework are pure speculation unless they are announced by this administration,” a White House spokesperson tells WIRED.

Key questions remain, for example, on how much control Oracle and ByteDance would each have on TikTok’s US data and algorithm. Trump’s Truth Social post suggests that he will meet with Xi again at the APEC Summit in South Korea in late October, meaning details could emerge after that.

On Thursday, during a joint press conference with UK prime minister Keir Starmer, Trump boasted that the US should receive a “tremendous fee plus” for brokering the TikTok deal. It’s not clear what fee he’s referring to—WIRED asked multiple White House officials, but none replied.

The White House also credited Vice President JD Vance—the top conduit between Silicon Valley billionaires and the West Wing—for playing a key role in the deal. A White House official told WIRED that Sean Cooksey, an adviser to Vance, was “at the forefront” of negotiations on behalf of the vice president.

US efforts to ban TikTok started during Trump’s first term in 2020. Months before he left office, Trump threatened to ban TikTok and another Chinese app WeChat. The Biden administration rescinded Trump’s executive orders on the topic but continued to scrutinize TikTok. The US congress eventually passed the Protecting Americans from Foreign Adversary Controlled Applications (PAFACA) Act in April 2024. This gave TikTok two options: divest from its Chinese ownership before January 19, 2025, or risk a federal ban.