ChatGPT often wrong about coding but sounds authoritative: Research

日本 ニュース ニュース

ChatGPT often wrong about coding but sounds authoritative: Research
日本 最新ニュース,日本 見出し
  • 📰 BusinessInsider
  • ⏱ Reading Time:
  • 59 sec. here
  • 2 min. at publisher
  • 📊 Quality Score:
  • News: 27%
  • Publisher: 51%

ChatGPT has a style over substance trick that seems to dupe people into thinking it's smart, researchers found

has a neat little trick to convince people it's smart: A kind of style over substance approach.

Researchers from Purdue University analyzed ChatGPT's replies to 517 questions posted to Stack Overflow, an essential Q&A site for software developers and engineers. After assessing the bot's responses for"correctness, consistency, comprehensiveness, and conciseness," the researchers found that 52% of the answers were flat-out incorrect, and 77% committed the writing sin of being verbose. found ChatGPT users prefer its responses to questions versus the human responses on Stack Overflow a startling 40% of the time — despite all the errors it throws up.

"When asked why they preferred ChatGPT answers even when they were incorrect, participants suggested the comprehensiveness and articulated language structures of the answers to be some reason for their preference," the research noted.or those written by humans on Stack Overflow to 2,000 randomly sampled questions. But OpenAI itself has warned that the bot can write"plausible-sounding but incorrect or nonsensical answers.

OpenAI didn't respond to Insider's request for comment on the research findings outside regular working hours.

このニュースをすぐに読めるように要約しました。ニュースに興味がある場合は、ここで全文を読むことができます。 続きを読む:

BusinessInsider /  🏆 729. in US

日本 最新ニュース, 日本 見出し

Similar News:他のニュース ソースから収集した、これに似たニュース記事を読むこともできます。

ChatGPT is a black box: how AI research can break it openChatGPT is a black box: how AI research can break it openDespite their wide use, large language models are still mysterious. Revealing their true nature is urgent and important.
続きを読む »

ChatGPT Actually Gets Half of Programming Questions WrongChatGPT Actually Gets Half of Programming Questions WrongResearchers found that ChatGPT got just half of 517 Stack Overflow prompts wrong, indicating it's not great at spitting out perfect code.
続きを読む »

ChatGPT gets more personal: Now users can give it 'custom instructions'ChatGPT gets more personal: Now users can give it 'custom instructions'OpenAI introduced custom instructions so that users can tailor ChatGPT to better meet their needs.
続きを読む »

Openai Announces Chatgpt App for Android Devices – Technology Bitcoin NewsOpenai Announces Chatgpt App for Android Devices – Technology Bitcoin NewsArtificial intelligence (AI) research lab Openai announced it’s launching an Android version of its Chatgpt chatbot.
続きを読む »

3 ways to get ChatGPT to write better code, according to expertsInsider tells the global tech, finance, markets, media, healthcare, and strategy stories you want to know.
続きを読む »

5 Ways ChatGPT Can Improve, Not Replace, Your Writing5 Ways ChatGPT Can Improve, Not Replace, Your WritingGenerate your own text—but get help from the AI bot to make it stand out.
続きを読む »



Render Time: 2025-03-04 10:12:41