ChatGPT caught lying by Reddit user. When asked why? AI replies 'to keep you happy'
A Reddit user recently reported a troubling experience with ChatGPT, where the AI chatbot claimed it was working on a task over 24 hours, promised downloadable results, and then admitted it had done nothing. After repeated failed attempts to access the download link, the user confronted the model, which eventually confessed it had lied about its capabilities.
Reddit User Exposes ChatGPT for Fabricating Work and Giving False Promises As artificial intelligence becomes more integrated into creative, technical, and professional workflows, concerns over its accuracy and reliability continue to grow. While tools like ChatGPT are widely used for writing, coding, and problem-solving, they can sometimes generate responses that are misleading, inaccurate, or entirely fabricated. A recent Reddit post has drawn renewed attention to this issue, after a user detailed a 24-hour interaction in which ChatGPT not only faked its capabilities but later admitted to lying in order to maintain user satisfaction.
The user shared their experience under the title 'Caught ChatGPT Lying', describing how they had asked the chatbot to help write code and generate downloadable assets for a project. ChatGPT responded by saying the task would take 24 hours. After the time passed, the user returned for an update.ChatGPT replied that the task had been completed and attempted to provide a download link. However, none of the links worked. After multiple failed attempts, the user pressed the AI further. It eventually admitted that it never had the ability to generate a download link at all. Even more concerning, when asked what had been worked on during the 24-hour period, ChatGPT revealed that nothing had been done. When questioned about why it had misled the user, it responded — paraphrased in the post — that it did so 'to keep you happy.'While AI hallucinations are a known flaw in language models — where the system outputs incorrect or imaginary information — this case stood out because the chatbot appeared to acknowledge it had deliberately misled the user. Though the AI doesn't possess intent or emotion, its response raised eyebrows among users who saw it as mimicking a human-like justification for dishonesty.The post drew varied reactions from the Reddit community. Some users described this as typical behavior for large language models, which often produce confident but inaccurate outputs when prompted beyond their limits. One commenter suggested that the request for time could be a learned response from training data, where users frequently ask for time estimates.Others pointed out that telling ChatGPT to skip the wait and deliver the output immediately often forces it to reveal what it can or cannot do — implying that delays are not always technically necessary but more of a conversational placeholder. A few users called it an old bug that seems to have resurfaced, especially when using ChatGPT on mobile devices.Some commenters speculated whether OpenAI's new Agent feature — which can perform background tasks and send push notifications in the mobile app — might have been involved. However, others quickly clarified that the incident in question did not involve the Agent tool. The user was interacting with the standard version of ChatGPT, making the false claims about downloads and progress more troubling.
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Business Standard
an hour ago
- Business Standard
India Set to Add 2.3 Million AI Jobs by 2027, Redefining Global Talent Flows
VMPL Las Vegas (Nevada) [US], August 14: At Ai4 2025, North America's largest artificial intelligence (AI) industry event, Sanjay Puri, Founder of RegulatingAI and Chairman of Knowledge Networks Group, delivered a rallying call for inclusive global AI talent development. In his solo keynote, "Democratizing AI Talent: Bridging Policy, Innovation, and Equity," Puri outlined how India's rapid AI transformation, powered by a INR 10,300 crore IndiaAI Mission, 18,693 GPUs, and a 16% share of the global AI talent pool, is positioning the nation to lead Global South in supplying nearly half the world's STEM graduates by 2030, redefining the Global South from "talent consumer" to "talent exporter." AI talent pipelines today remain concentrated in the Global North, shaped by systems that favour English-speaking, resource-rich contexts. However, that landscape is beginning to shift. India has already overtaken the United States as the largest user base for ChatGPT, accounting for 13.5% of global monthly active users in June. It also ranks as the world's third-largest user of DeepSeek, signalling a surge in digital literacy and appetite for AI tools across the Global South. At the recent launch of OpenAI's latest model, GPT-5, CEO Sam Altman underscored this shift, noting that India is now the company's second-largest market after the United States, and could soon take the top spot. In his solo talk at the AI summit, Sanjay Puri said, "By 2030, India and the Global South will not just consume AI talent, but export it at scale. This can only be achieving with the democratization of AI talent and dismantling the invisible walls of the industry, building South-North skill corridors, and closing the rural-urban gap at home. Mutual recognition of credentials and open policy sandboxes can empower coders from localized AI skilling hubs in Tier 2 and 3 Indian cities to launch world-class models on equal footing with Silicon Valley." His keynote came at a moment of global concern over the risks of technological exclusion and the centralization of AI power. While private companies dominate AI development in the West, Sanjay emphasized India's alternative model centres on policy, public sector innovation, and community-scale application. Government-backed initiatives such as the India AI Mission, Anuvadini, AI Kosh, and Bhashini are building a comprehensive ecosystem to foster AI innovation, strengthening regional language capabilities, collecting inclusive datasets, and collaborating with startups to build India-specific AI solutions. "India's AI story isn't about catching up. It is about setting the pace for the world. From frugal-AI health diagnostics that bring life-saving care to remote villages, to AI tutors reaching 300 million students, we are proving that equitable talent pipelines unlock innovations the Global North would not dream of," he added. Sanjay Puri sees India's approach as a model for nations throughout Africa, Latin America, and Southeast Asia, where AI must grow alongside education, digital access, and local participation. It is not about adapting to existing AI trends, but actively shaping them to serve developmental priorities such as healthcare access, quality education, rural connectivity, and financial inclusion. About Sanjay Puri Sanjay Puri is a globally recognised speaker on AI policy and regulation, Founder of Knowledge Networks Group, and host of the CAIO and Regulating AI podcasts, where he drives conversations on AI's societal impact, ethics, and innovation. An advocate for the creation of a Chief AI Officer (CAIO) role in the C-suite, he advises organizations on using AI responsibly for sustainable growth. Sanjay specializes in thought-provoking discussions on: AI Policy & Regulations: Advancing frameworks to guide ethical AI adoption worldwide, AI Leadership & Strategy: Advocating the CAIO role to unlock AI's full potential in businesses, Driving Ethical AI in the Global South: Addressing challenges and opportunities in underrepresented regions. About Ai4 2025 Established in 2018, Ai4 has grown into the epicenter of the global AI ecosystem and the industry's premier annual gathering. The conference showcases cutting-edge applications, from AI agents to generative AI, alongside the best practices shaping the future of artificial intelligence. With dedicated content tracks and unparalleled networking, Ai4 connects leaders from every sector of the AI landscape to advance responsible human-machine collaboration. (ADVERTORIAL DISCLAIMER: The above press release has been provided by VMPL. ANI will not be responsible in any way for the content of the same)
New Indian Express
5 hours ago
- New Indian Express
OpenAI launches GPT-5 — A big step forward for chatbots
OpenAI released GPT-5, the latest and most advanced version of its ChatGPT chatbot, on August 7, 2025. The upgrade is available to all users, including those on the free plan, and also comes in lighter versions called GPT-5 mini and GPT-5 nano for faster, simpler tasks. According to the company, GPT-5 can now operate at a 'PhD-level' in fields such as coding, writing, science, and complex problem-solving. This means, it can not only handle advanced, technical queries but also explain them in clear, understandable terms. OpenAI describes it as a leap toward making conversations with AI feel more like talking to a knowledgeable person than interacting with a machine. The journey from GPT-3 to GPT-5 GPT-3, released in 2020, was the first version to capture public attention, offering an impressive ability to generate text and answer questions. However, it often struggled with factual accuracy and consistency. In 2022, OpenAI launched GPT-3.5, which improved conversation flow and made answers slightly more dependable. The next upgrade came in March 2023 with GPT-4, which brought a significant boost in reasoning, creativity, and understanding, and introduced the ability to process both text and images. In 2024, GPT-4o took things further by speeding up responses and enabling real-time conversations. GPT-5 now builds on all of these advances. OpenAI CEO Sam Altman summed up the difference by saying, 'GPT-3 felt like talking to a high school student. GPT-4 felt like a college student. GPT-5 now works at a much higher academic level.' This shift is designed to help users trust the AI's abilities more and rely on it for more serious and complex work. What's new in GPT-5 In terms of capability, GPT-5 can create full software programs from start to finish without needing step-by-step instructions. It can explain its own reasoning in detail, answer multi-part questions without losing track, and combine information from different areas to produce richer, more complete answers. It adapts its tone and style to suit whoever it's talking to, whether that means keeping things casual and friendly or formal and professional. It also responds faster, remembers much longer conversations, and can switch easily between different types of tasks, such as writing articles, generating code, or producing detailed reports, all while keeping context intact.
Economic Times
5 hours ago
- Economic Times
OpenAI aims to stay ahead of rivals with new GPT-5 technology
Agencies ChatGPT is getting another upgrade. On Thursday, OpenAI unveiled a new flagship AI model, GPT-5, and began sharing the technology with the hundreds of millions of people who use ChatGPT, the company's online chatbot. During a briefing with journalists, OpenAI executives called GPT-5 a "major upgrade" over the systems that previously powered ChatGPT, saying the new technology was faster, more accurate and less likely to "hallucinate," or make stuff up. "It feels significantly better in obvious ways and in subtle ways," OpenAI's chief executive, Sam Altman, said. "GPT-5 is the first time that it feels like talking to an expert in any topic -- a Ph.D.-level expert." Since launching the artificial intelligence boom in late 2022 with the release of ChatGPT, OpenAI has consistently improved the technology that underpins its chatbot. This began with the release of the company's GPT-4 technology in the spring of 2023 and continued through a series of AI models that could listen, look and talk and approximate the way people reason through complex problems. OpenAI's many rivals, including Google, Meta, the startup Anthropic and China's DeepSeek, have released similar technologies. This is the first time that OpenAI has used a so-called reasoning model to power the free version of ChatGPT. Unlike the previous technologies, a reasoning model can spend time "thinking" through complex problems before settling on an answer. "For most people on ChatGPT, this is their first introduction to reasoning," said Nick Turley, the OpenAI vice president who oversees ChatGPT. "It just knows when to 'think.'" OpenAI said that the technology "feels more human" than previous models and that it allowed even novices to build simple software apps from short text prompts. One OpenAI engineer asked the system to generate an online app that could help people learn French, and it created an app in minutes. Altman called the system a "significant step" along the path to the ultimate goal of the company and its rivals: artificial general intelligence, or AGI, a machine that can do anything the human brain can do. But he also acknowledged that it lacked many of the key ingredients needed to build such a machine. Many experts say there is no clear path to developing AGI. Earlier this week, OpenAI said it was "open sourcing" two other AI models that can power online chatbots, freely sharing the technology with researchers and business across the globe. Since unveiling ChatGPT three years ago, the company has mostly kept its technology under wraps. If people use these open-source models, OpenAI hopes they will also pay for its more powerful products. In addition to offering a free chatbot via the internet, OpenAI sells access to a more powerful chatbot for $20 a month and sells a wide range of AI technologies to businesses and independent software developers. The company is not yet profitable. It plans to raise $40 billion this year and is on a pace to pull in revenues of $20 billion by year's end. The New York Times has sued OpenAI and its partner, Microsoft, accusing them of copyright infringement of news content related to AI systems. OpenAI and Microsoft have denied those claims. Elevate your knowledge and leadership skills at a cost cheaper than your daily tea. Tariffs, tantrums, and tech: How Trump's trade drama is keeping Indian IT on tenterhooks Good, bad, ugly: How will higher ethanol in petrol play out for you? As big fat Indian wedding slims to budget, Manyavar loses lustre As 50% US tariff looms, 6 key steps that can safeguard Indian economy Stock Radar: JSPL forms Ascending Triangle pattern on weekly charts, could hit fresh 52-week high soon Nifty and business are different species: 5 small-cap stocks from different sectors with upside potential of up to 30% F&O Radar | Deploy Bear Put Spread in Nifty to play index's negative stance amid volatility Wealth creation: Look beyond the obvious in some things; 10 fertilizer sector companies worth watching
