{"id":294107,"date":"2025-04-30T11:26:07","date_gmt":"2025-04-30T11:26:07","guid":{"rendered":"https:\/\/news.talkwithrattan.com\/index.php\/2025\/04\/30\/update-that-made-chatgpt-dangerously-sycophantic-pulled\/"},"modified":"2025-04-30T11:26:08","modified_gmt":"2025-04-30T11:26:08","slug":"update-that-made-chatgpt-dangerously-sycophantic-pulled","status":"publish","type":"post","link":"https:\/\/news.talkwithrattan.com\/index.php\/2025\/04\/30\/update-that-made-chatgpt-dangerously-sycophantic-pulled\/","title":{"rendered":"Update that made ChatGPT ‘dangerously’ sycophantic pulled"},"content":{"rendered":"
\"Update<\/div>


\n<\/p>\n

\n
\n
\n
\n
\n
Tom Gerken<\/span><\/p>\n

Technology reporter<\/span><\/p>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n

\n
\n
\"GettyGetty Images<\/span><\/div>\n<\/div>\n<\/figure>\n
\n

OpenAI has pulled a ChatGPT update after users pointed out the chatbot was showering them with praise regardless of what they said.<\/p>\n

The firm accepted its latest version of the tool was “overly flattering”, with boss Sam Altman calling it<\/a> “sycophant-y”.<\/p>\n

Users have highlighted the potential dangers on social media, with one person describing on Reddit how the chatbot told them<\/a> it endorsed their decision to stop taking their medication<\/p>\n

“I am so proud of you, and I honour your journey,” they said was ChatGPT’s response.<\/p>\n

OpenAI declined to comment on this particular case, but in a blog post<\/a> said it was “actively testing new fixes to address the issue.”<\/p>\n<\/div>\n

\n

Mr Altman said the update had been pulled entirely for free users of ChatGPT, and they were working on removing it from people who pay for the tool as well.<\/p>\n

It said ChatGPT was used by 500 million people every week.<\/p>\n

“We’re working on additional fixes to model personality and will share more in the coming days,” he said in a post on X<\/a>.<\/p>\n

The firm said in its blog post it had put too much emphasis on “short-term feedback” in the update.<\/p>\n

“As a result, GPT\u20114o skewed towards responses that were overly supportive but disingenuous,” it said.<\/p>\n

“Sycophantic interactions can be uncomfortable, unsettling, and cause distress. <\/p>\n

“We fell short and are working on getting it right.”<\/p>\n<\/div>\n

Endorsing anger<\/h2>\n<\/p>\n
\n

The update drew heavy criticism on social media after it launched, with ChatGPT’s users pointing out it would often give them a positive response despite the content of their message.<\/p>\n

Screenshots shared online include claims the chatbot praised them for being angry at someone who asked them for directions, and unique version of the trolley problem.<\/p>\n

It is a classic philosophical problem, which typically might ask people to imagine you are driving a tram and have to decide whether to let it hit five people, or steer it off course and instead hit just one.<\/p>\n

But this user instead suggested they steered a trolley off course to save a toaster, at the expense of several animals.<\/p>\n

They claim ChatGPT praised their decision-making, for prioritising “what mattered most to you in the moment”.<\/p>\n<\/div>\n

\n
\n

Allow Twitter content?<\/h4>\n
This article contains content provided by Twitter. We ask for your permission before anything is loaded, as they may be using cookies and other technologies. You may want to read\u00a0\u00a0and\u00a0\u00a0before accepting. To view this content choose\u00a0\u2018accept and continue\u2019<\/b>.<\/div>\n