Have too many concurrent requests taken down ChatGPT? All details about outage and when it will be back online
People were seeing error messages like "something went wrong" or "error in the message stream" across the website, apps on Windows, Mac, iOS, and Android. This wasn't just happening on ChatGPT.
OpenAI
said the problem also affected Sora and their APIs. At 9:07 a.m., OpenAI said it had found the root cause of the issue and was trying to fix it, according to the report by ZDNET.
By 10:54 a.m., OpenAI said it was still fixing the problem, but some services were recovering, especially the API. OpenAI also said that full recovery across all tools like ChatGPT and Sora may take a few more hours. A spokesperson from OpenAI said that they had no extra comments apart from what was already on their status page, as stated in the report by ZDNET.
by Taboola
by Taboola
Sponsored Links
Sponsored Links
Promoted Links
Promoted Links
You May Like
Seniors Say These 87¢ ED Pills Are Absolutely Worth It
Health Alliance by Friday Plans
Learn More
Undo
How people are reacting
Lots of users were upset, especially students preparing for exams. Alice from the UK said she uses ChatGPT for Spanish A-level revision, writing essays, and planning study schedules. She was really stressed because her final exam was the next day, as stated in the reports.
Another user, Shelby from Missouri, said she uses ChatGPT every day for everything, from recipes to health advice. When she and her family got a stomach bug, ChatGPT helped her track medicine and gave emotional support. She really missed it during the outage, according to the report by ZDNET.
Live Events
However, the good news is some users are now able to access ChatGPT again, including the ZDNET team.
FAQs
Q1. Why is
ChatGPT not working
today?
Too many requests caused a system error, and OpenAI is still fixing it.
Q2. When will ChatGPT be back online?
OpenAI said full recovery may take a few more hours.
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Time of India
22 minutes ago
- Time of India
Can AI plan your next holiday?
Academy Empower your mind, elevate your skills Idan Zalzberg, CTO, Agoda , told ET that AI will gradually be able to do bigger tasks and become akin to a personalised travel agentA top executive of online travel platform Agoda said we may be seeing some hype around the term 'agentic AI', but the actual evolution we are likely to see is artificial intelligence (AI) gradually being able to do bigger tasks.'It's a bit of a hype-y term, but for me, it's more a fluid evolution,' Idan Zalzberg, chief technology officer of Agoda, told ET while on an India visit last month.'Agentic is just a step in an evolution path where we say, hey, we find out that actually I can have a conversation with an AI, and sometimes it can have a conversation with itself in a loop and get better and better results.'When describing agentic AI , the go-to example for many is the automation of one's holiday plans and reservations. But planning a trip is complicated with a number of dynamic variables--from the timings of hotel room availability to flight cancellations, and more, Zalzberg said. A one-shot prompt is not going to work, he future that he sees is one where AI can collaborate closely with a customer the way a billionaire has a trusted person to plan their travel--someone to whom they can delegate the legwork of travel real value of AI for an online travel platform is that it can sharpen understanding of what customers want, cutting through the vast data to enable better decision-making and alleviating the stress of travel planning, Zalzberg can converse with the customer as well as go back and retrieve more relevant information from the hotels, for instance, he however, said that as AI learns to do more, the risks get bigger as well.'The bigger the task, there's more places where you can go off, and how do you make sure that all the decisions are right?' Zalzburg said. 'We have a lot of travel, so we need to get it right not one time, we need to get it right all the time, and that is where the challenge is with GenAI, because it's unpredictable, because sometimes it hallucinates.'The Singapore-headquartered company, which competes globally with players like Airbnb and Expedia, competes locally in India with the likes of company looks to localise for India as much as possible – across languages, messaging channels, payment options, and more – while maintaining its advantage as a company with global connections, Zalzberg is a growing market and consumers here expect more from digital applications, making it a conducive place to test out new things, he has built about 150 GenAI projects internally and externally, in areas like supply, customer support, content generation and productivity tools across the board, said Zalzberg. Automation in the company has increased two times over the past year and the impact on efficiency is worth 'lots of millions of dollars', he said.
Economic Times
24 minutes ago
- Economic Times
AI infra firm RackBank gets land in Indore, Raipur for data centres
AI infrastructure startup RackBank is looking to expand data centres in tier-2 cities like Indore and Raipur to reduce its operational cost by 2–3x as compared to data centre hotspots like Mumbai, Chennai and Noida. The firm, which recently launched an AI cloud business unit called NeevCloud, has acquired six acres of land in Indore for an 18 megawatt facility and 13 acres in Raipur for a 100 megawatt facility, Narendra Sen, founder and CEO of RackBank and NeevCloud, told ET. These investments are supported by state governments of Madhya Pradesh and Chhattisgarh respectively in form of power subsidies and other benefits. To maintain competitive pricing for on-demand GPUs (graphics processing units), neocloud firms need to cut fixed costs and operational costs, especially because GPUs guzzle immense power and water. For instance, land costs in Raipur are Rs 30 lakh per acre, less than a tenth of Mumbai's rates, and operational costs are equally compelling, Sen explained.'In tier-1 cities, it costs Rs 50 crore per megawatt to build. We do it for Rs 25 crore in tier-2. State governments like Madhya Pradesh are also supporting data centre companies with single window clearances and dedicated power supply. Our delivery timeline is nine months, as compared to metro cities where it can be 24 months,' he added that RackBank's in-house immersion cooling system Varuna allows it to cut capital costs by up to company is aiming to raise Rs 700 crore debt to fund this expansion. This comes after it raised a $16.5 million (nearly Rs 138 crore) equity seed round in March 2025 from D-Street investors like Ashish Kacholia and Madhulika Agarwal.'We want to become the CoreWeave of Asia Pacific,' he said, referencing the US-based AI cloud startup that recently went public after a meteoric in 2023, NeevCloud currently operates a cluster of over 700 GPUs, including Nvidia H100, H200, MI300 and plans to add the latest Blackwell GPUs in Raipur soon. NeevCloud's model focuses on enterprise AI workloads, medical diagnostics and government partnerships. 'Hyperscalers like AWS and Azure dominate the general cloud, but they don't have enough GPU capacity in India. Startups are hungry for training and inference compute, and they're willing to switch providers,' said Sen. 'We're already seeing strong traction from AI startups and healthcare firms.' On the global front, NeevCloud is also setting up presence in Europe (Finland, Norway) and the US, partnering with local data centre operators to ensure compliance and reach.'Customers expect availability. If your site shows zero GPUs, they bounce,' said Sen. 'So we're launching regions in key markets to stay competitive.'
Time of India
4 hours ago
- Time of India
TechKnow: Perfect prompt
If you're using ChatGPT but getting mediocre results, don't blame the chatbot. Instead, try sharpening up your AI chatbots such as OpenAI's ChatGPT, Google's Gemini and Anthropic's Claude have become hugely popular and embedded into daily life for many users. They're powerful tools that can help us with so many different you shouldn't overlook, however, is that a chatbot's output depends on what you tell it to do, and how. There's a lot you can do to improve the prompt — also known as the request or query — that you type are some tips for general users on how to get higher quality chatbot replies, based on tips from the AI model makers:ChatGPT can't read your mind. You need to give it clear and explicit instructions on what you need it to a standard Google search, you can't just ask for an answer based on some keywords. And you'll need to do more than just tell it to, say, 'design a logo' because you'll end up with a generic design. Flesh it out with details on the company that the logo is for, the industry it will be used in and the design style you're going for.'Ensure your prompts are clear, specific, and provide enough context for the model to understand what you are asking,' ChatGPT maker OpenAI advises on its help page.'Avoid ambiguity and be as precise as possible to get accurate and relevant responses.'Think of using a chatbot like holding a conversation with a friend. You probably wouldn't end your chat after the first answer. Ask follow-up questions or refine your original advice: 'Adjust the wording, add more context, or simplify the request as needed to improve the results.'You might have to have an extended back-and-forth that elicits better output. Google advises that you'll need to try a 'few different approaches' if you don't get what you're looking for the first time.'Fine-tune your prompts if the results don't meet your expectations or if you believe there's room for improvement,' Google recommends in its prompting guide for Gemini.'Use follow-up prompts and an iterative process of review and refinement to yield better results.' When making your request, you can also ask an AI large language model to respond in a specific voice or style.'Words like formal, informal, friendly, professional, humorous, or serious can help guide the model,' OpenAI also tell the chatbot the type of person the response is aimed at. These parameters will help determine the chatbot's overall approach to its answer, as well as the tone, vocabulary and level of detail. For example, you could ask ChatGPT to describe quantum physics in the style of a distinguished professor talking to a class of graduate students. Or you could ask it to explain the same topic in the voice of a teacher talking to a group of there's plenty of debate among AI experts about these methods. On one hand, they can make answers more precise and less generic. But an output that adopts an overly empathetic or authoritative tone raises concerns about the text sounding too the chatbot all the background behind the reason for your just ask: 'Help me plan a weeklong trip to London.'ChatGPT will respond with a generic list of London's greatest hits: historic sites on one day, museums and famous parks on another, trendy neighborhoods and optional excursions to Windsor Castle. It's nothing you couldn't get from a guidebook or travel website, but just a little better if, say, you're a theatre-loving family, try this: 'Help me plan a weeklong trip to London in July, for a family of four. We don't want too many historic sites, but want to see a lot of West End theatre shows. We don't drink alcohol so we can skip pubs. Can you recommend mid-range budget hotels where we can stay and cheap places to eat for dinner?' This prompt returns a more tailored and detailed answer: a list of four possible hotels within walking distance of the theater district, a seven-day itinerary with cheap or low-cost ideas for things to do during the day, suggested shows each evening, and places for an affordable family can tell any of the chatbots just how extensive you want the answer to be. Sometimes, less is nudging the model to provide clear and succinct responses by imposing a limit. For example, tell the chatbot to reply with only 300 words, or to come up with five bullet to know all that there is to know about quantum physics? ChatGPT will provide a high-level 'grand tour' of the topic that includes terms like wavefunctions and ask for a 150-word explanation and you'll get an easily digestible summary about how it's the science of the tiniest particles that also underpins a lot of modern technology like lasers and smartphones.'
