OpenAI Introduces ChatGPT Search, Now Available to ChatGPT Plus Users and Those on Waiting List; Roll Out to Other Users Will Happen Over the Coming and Months
From OpenAI:
You can get fast, timely answers with links to relevant web sources, which you would have previously needed to go to a search engine for. This blends the benefits of a natural language interface with the value of up-to-date sports scores, news, stock quotes, and more.
ChatGPT will choose to search the web based on what you ask, or you can manually choose to search by clicking the web search icon.
[Clip]
Search will be available at chatgpt.com, as well as on our desktop and mobile apps. All ChatGPT Plus and Team users, as well as SearchGPT waitlist users, will have access today. Enterprise and Edu users will get access in the next few weeks. We’ll roll out to all Free users over the coming months.
[Clip]
The search model is a fine-tuned version of GPT-4o, post-trained using novel synthetic data generation techniques, including distilling outputs from OpenAI o1-preview. ChatGPT search leverages third-party search providers, as well as content provided directly by our partners, to provide the information users are looking for. Learn more here.
Learn More, Read the Complete Post
More From The Verge:
Prior to this update, ChatGPT’s knowledge was limited to a cutoff between 2021 and 2023 depending on the model. OpenAI spokesperson Niko Felix said even with live search active, the company will continue to refresh its training data to “ensure our users always have access to the latest advancements” but it is “distinct” from the training of the company’s models.
[Clip]
As for hallucinations — like when Google’s AI Overviews infamously told users to put glue on their pizza — Fry believes that ChatGPT search is going to “increase factual accuracy overall.”
“Some amount of hallucinations come from just really not having access to the latest information. And so now that it has access to the latest information, it actually helps it make better decisions around what is the true, factual answer,” Fry said. When pressed about the fact that edge cases exist and mistakes are inevitable, he said “we’ll try to be transparent” if it does trip up.
Learn More, Read the Complete Article
From WIRED:
I was quite anxious about the ramifications when I first saw my writing for WIRED referenced by ChatGPT back in 2023. After a few hours of testing a prelaunch version of ChatGPT’s new AI search, it’s clear to me that OpenAI has made significant progress since its initial messy foray into web browsing, with more interactive elements and clearer attribution of its sources. I could see a subset of early adopters really latching on to the new ChatGPT search.
Taking that into account, the product needs improvements before it’s able to truly compete with the dominance of Google for essential search experiences, such as online shopping. ChatGPT also makes some of the same mistakes as other AI search tools, such as hallucinating and citing incorrect information.
Learn More, Read the Complete Article
It’s a positive development that ChatGPT will now be able to retrieve information from these reputable online sources and generate answers based on them, says Suzan Verberne, a professor of natural-language processing at Leiden University, who has studied information retrieval. It also allows users to ask follow-up questions.
But despite the enhanced ability to search the web and cross-check sources, the tool is not immune from the persistent tendency of AI language models to make things up or get it wrong. When MIT Technology Review tested the new search function and asked it for vacation destination ideas, ChatGPT suggested “luxury European destinations” such as Japan, Dubai, the Caribbean islands, Bali, the Seychelles, and Thailand. It offered as a source an article from the Times, a British newspaper, which listed these locations as well as those in Europe as luxury holiday options.
“Especially when you ask about untrue facts or events that never happened, the engine might still try to formulate a plausible response that is not necessarily correct,” says Verberne. There is also a risk that misinformation might seep into ChatGPT’s answers from the internet if the company has not filtered its sources well enough, she adds.
Filed under: Companies (Publishers/Vendors), Data Files, News, Patrons and Users
About Gary Price
Gary Price (gprice@gmail.com) is a librarian, writer, consultant, and frequent conference speaker based in the Washington D.C. metro area. He earned his MLIS degree from Wayne State University in Detroit. Price has won several awards including the SLA Innovations in Technology Award and Alumnus of the Year from the Wayne St. University Library and Information Science Program. From 2006-2009 he was Director of Online Information Services at Ask.com.
ADVERTISEMENT
Archives
ADVERTISEMENT