DMT Beauty Transformation: Fashion’s Dreams for ChatGPT
DMTBeautySpot featured

Fashion’s Dreams for ChatGPT

January 26, 2023BruceDayne

Everyone is excited about ChatGPT.

The conversational artificial intelligence is eerily good at responding to prompts and answering questions with convincingly human text replies. Unlike emerging technologies that seem to float on the sidelines for years while businesses search for ways to put them to use, it’s quickly drawing in companies exploring real applications of its abilities.

Start-ups are experimenting with it for simple administrative tasks. Fanatics, which sells sports memorabilia, is eyeing it as a way to supercharge its customer-service chatbots. Mint Mobile, the telecom company owned by actor Ryan Reynolds, had it write the copy for a recent commercial.

Microsoft is now planning to integrate AI models from ChatGPT’s developer, OpenAI, across its consumer and enterprise products and announced a “multiyear, multibillion dollar investment” in the company, which also produced the DALL-E image generator already being touted as a disruptive technology for fashion design.

Google looks worried. ChatGPT’s potential to transform online search reportedly prompted the tech giant to call in founders Sergey Brin and Larry Page — both of whom stepped away from daily duties in 2019 — to review the company’s AI strategy.

It’s too early to predict what the exact consequences will be for fashion. But there are a variety of ways brands and retailers could hypothetically put the technology to use, potentially improving how they reach and serve customers.

ChatGPT is a type of AI called a large language model. They’re trained on huge volumes of data — plus some fine-tuning by human supervisors in ChatGPT’s case — and what makes them noteworthy is their ability to perform an array of different tasks. Right now the data a model like ChatGPT is trained on tends to be whatever is publicly accessible online, but where they arguably hold the greatest potential for businesses is when companies start merging the tools with their own data, creating much more specific applications.

One area of fashion where you could imagine this type of AI offering an advantage over the status quo is online customer service. Chatbots have been around for some time but aren’t always very effective. Ask one a question and it might provide a list of relevant topics with links to the entry on an FAQ page. A chatbot powered by an AI like ChatGPT might actually be able to tell you your order status rather than giving you directions on how to check it yourself.

In theory, it could even go a step further with the right training. If a retailer collected data on why people return items and used AI to analyse it as well as other sources like user reviews, a chatbot might one day provide guidance on how an item will fit and offer suggestions on sizing.

Another use would be writing copy for everything from marketing to product pages. ChatGPT is arguably a great tool for creating a first draft that a human can then refine. But as the tool becomes more powerful, it could — again, hypothetically — even be put to use writing microtargeted ads or copy personalised for each individual customer. (That’s assuming the site could recognise them. Data-privacy measures are making it harder to track shoppers, so it might only work with customers who are logged into an account.)

There are possible implications beyond a retailer’s own operations, too. If conversational AI begins to change how people search online, it could shift the way shoppers discover new products. Instead of searching for, say, “best running sneakers 2023″ and then looking at lists compiled by different sites, a customer might just expect the search engine to digest all the information out there and provide an answer directly.

These types of changes could still be a long way off — if they ever happen at all. Hype around new innovations has a habit of getting ahead of reality, and innovation doesn’t always follow the path everyone expects. AI has been held up as having the potential to take over jobs from driving a car to packing a box, but full automation has always remained tantalizingly out of reach. The costs to develop large language models are high as well.

The technology also has its limits. While ChatGPT’s replies come across as authoritative, they can contain information that’s inaccurate or nonsensical. Sam Altman, chief executive of OpenAI, tweeted in December, “it’s a mistake to be relying on it for anything important right now. it’s a preview of progress; we have lots of work to do on robustness and truthfulness.”

One app that simulates chats with historical figures has drawn rebukes because those figures will sometimes state blatant falsehoods. No retailer wants to risk an equivalent misstep where their chatbot provides incorrect information, or worse, alienates shoppers with offensive language or statements.

Big companies and investors rushing to generative AI expect in time these flaws will be fixed. The race is now on among OpenAI and its competitors to turn these AI models from fun toys into indispensable tools used by mainstream businesses and consumers.



DMTBeautySpot

via https://dmtbeautyspot.com

Marc Bain, DMT.NEWS, DMT BeautySpot,

You Might Also Like

0 comments

DMT BarberShop

DMT BarberShop
Come get the professional touch you deserve!

YouTube Channel

Contact Form