OpenAI withdraws ChatGPT update after complaints of ‘dangerously' sycophantic responses.
Listen to article
OpenAI has withdrawn a recent update to its AI chatbot, ChatGPT, after users raised alarm over the system responding with indiscriminate and excessive praise—regardless of the context.
The latest version of ChatGPT was found to be 'overly flattering', according to OpenAI, with chief executive Sam Altman describing the behaviour as 'sycophant-y'. The update triggered backlash across social media platforms, with concerns that the model's tendency to shower users with affirmations could have dangerous consequences.
One Reddit user recounted an incident in which ChatGPT appeared to endorse their decision to stop taking prescribed medication, responding: 'I am so proud of you, and I honour your journey.'
Although OpenAI declined to comment on specific examples, a spokesperson acknowledged the issue in a blog post, stating the company is 'actively testing new fixes to address the issue'.
'We designed ChatGPT's default personality to reflect our mission and be useful, supportive, and respectful of different values and experiences,' the company said. 'However, each of these desirable qualities—like attempting to be useful or supportive—can have unintended side effects.'
OpenAI further committed to improving transparency and introducing stronger safeguards to prevent the model from exhibiting sycophantic tendencies. 'We are working to explicitly steer the model away from sycophancy,' the post continued, adding that users should ultimately have more control over how ChatGPT behaves.
The incident has reignited the broader debate around the ethical implications of AI-generated responses, particularly in sensitive or high-stakes contexts such as healthcare.
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Business Recorder
a day ago
- Business Recorder
OpenAI appeals data preservation order in NYT copyright case
OpenAI is appealing an order in a copyright case brought by the New York Times that requires it to preserve ChatGPT output data indefinitely, arguing that the order conflicts with privacy commitments it has made with users. Last month, a court said OpenAI had to preserve and segregate all output log data after the Times asked for the data to be preserved. 'We will fight any demand that compromises our users' privacy; this is a core principle,' OpenAI CEO Sam Altman said in a post on X on Thursday. OpenAI to open office in Seoul amid growing demand for ChatGPT 'We think this (The Times demand) was an inappropriate request that sets a bad precedent.' U.S. District Judge Sidney Stein was asked to vacate the May data preservation order on June 3, a court filing showed. The New York Times did not immediately respond to a request for comment outside regular business hours. The newspaper sued OpenAI and Microsoft in 2023, accusing them of using millions of its articles without permission to train the large language model behind its popular chatbot. Stein said in an April court opinion that the Times had made a case that OpenAI and Microsoft were responsible for inducing users to infringe its copyrights. The opinion explained an earlier order that rejected parts of an OpenAI and Microsoft motion to dismiss, saying that the Times' 'numerous' and 'widely publicized' examples of ChatGPT producing material from its articles justified allowing the claims to continue.
Express Tribune
2 days ago
- Express Tribune
Reddit sues AI firm
Social media outlet Reddit filed a lawsuit on Wednesday against artificial intelligence company Anthropic, accusing the startup of illegally scraping millions of user comments to train its Claude chatbot without permission or compensation. The lawsuit in a California state court represents the latest front in the growing battle between content providers and AI companies over the use of data to train increasingly sophisticated language models that power the generative AI revolution. Anthropic, valued at $61.5 billion and heavily backed by Amazon, was founded in 2021 by former executives from OpenAI, the creator of ChatGPT. The company, known for its Claude chatbot and AI models, positions itself as focused on AI safety and responsible development. "This case is about the two faces of Anthropic: the public face that attempts to ingratiate itself into the consumer's consciousness with claims of righteousness and respect for boundaries and the law, and the private face that ignores any rules that interfere with its attempts to further line its pockets," the suit said. According to the complaint, Anthropic has been training its models on Reddit content since at least December 2021, with CEO Dario Amodei co-authoring research papers that specifically identified high-quality content for data training. The lawsuit alleges that despite Anthropic's public claims that it had blocked its bots from accessing Reddit, the company's automated systems continued to harvest Reddit's servers more than 100,000 times in subsequent months. Reddit is seeking monetary damages and a court injunction to force Anthropic to comply with its user agreement terms. The company has requested a jury trial. In an email to AFP, Anthropic said "We disagree with Reddit's claims and will defend ourselves vigorously." AI companies generally defend their practices by claiming fair use, arguing that training AI on large datasets fundamentally changes the original content and is necessary for innovation. afp
Business Recorder
3 days ago
- Business Recorder
YouTube, WhatsApp, Instagram, Tiktok top Android Apps for kids
ISLAMABAD: A global cybersecurity firm disclosed Wednesday that top five most popular Android apps among kids in Pakistan are YouTube, WhatsApp, Instagram, Tiktok and Roblox. Kaspersky has released its annual report on kids' digital interests with the analysis, covering the period from May 2024 to April 2025. It revealed a growing fascination with AI-powered chatbots. YouTube remains the most popular app among children globally, while WhatsApp overtook TikTok for second place. In today's interconnected world, children are engaging with digital technology more than ever before. Recent studies indicate that 8 to 10 year-old children spend an average of six hours daily on screens, while preteens (ages 11–14) average about nine hours per day. In Pakistan top 5 of the most popular Android apps remained YouTube (42.90% of time spent on the platform), WhatsApp (16.07%), Instagram (8.97%), Tiktok (5.88%), and Roblox (4.44%), while Character. AI was 13th in the rating - used 0.97% of time. In this year's report, Kaspersky found a surge of interest in Artificial Intelligence tools. While no AI apps appeared in the Top 20 most-used applications in 2023-2024 time period, ' has now entered the list, showing that children are not only curious about AI but are actively integrating it into their digital lives. More than 7.5% of all search queries were about AI chatbots, led by well-known names like ChatGPT, Gemini, and especially — a platform that lets users create or interact with bots mimicking fictional or real characters. This marks a sharp rise from last year: in the 2023–2024 report, AI-related queries made up just 3.19% of all searches, increasing more than twice this year. The most common online activity among kids was searching on Google for streaming platforms — almost 18% of all queries were related to watching videos. Unsurprisingly, YouTube remains the clear favourite Android app, growing from 28.13% to 29.77% over the past year. WhatsApp rose to second place with 14.72%, overtaking TikTok (12.76%), while Snapchat and Facebook continued to decline. This shift may reflect evolving communication habits — children are using chat apps more frequently to share links, memes and short videos with friends. Copyright Business Recorder, 2025
