ChatGPT users face widespread outage across the world, including India
In the last few hours, more than 800 people from India have complained that they are unable to use the AI chatbot. Following user reports, OpenAI has also updated its status to reflect that the AI chatbot they are 'currently experiencing issues'.
The Sam Altman-led AI startup said that 'some users are experiencing elevated error rates and latency across APIs, ChatGPT, and Sora.'
At the time of writing, paid ChatGPT subscribers who tried using its text-to-video generation model, Sor, are also facing problems for the last 37 minutes.
Developing story…
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Time of India
24 minutes ago
- Time of India
How to switch back to GPT-4 if you don't like ChatGPT-5 and why users are doing it?
OpenAI CEO Sam Altman on Wednesday announced a range of updates to ChatGPT, offering users greater control over the company's latest GPT-5 model and introducing expanded model options for paid subscribers. However, it seems to have not sat well with users, as since the launch of ChatGPT-5, many users have been flooding social media with complaints about the latest update. From slower response times to less engaging and more robotic answers, frustration has been mounting among users. Hence, as the disappointment grows, a significant number are eager to switch back to GPT-4, which they find more reliable and conversational. If you are among those unhappy with the latest update, here's how you can revert to GPT-4 and why so many are making the switch back. Why are so many disappointed with ChatGPT-5? Unlike its predecessors, a backlash has sparked a fresh debate over the psychological attachments some users form with the chatbots trained to push their emotional buttons. While some Reddit users dismissed complaints about GPT-5 as evidence of an unhealthy dependence on an AI companion. OpenAI's GPT-5 was intended to be a groundbreaking upgrade to its widely used and intelligent chatbot. However, for many users, last Thursday's release felt like a disappointing step backwards, as the new ChatGPT appeared to lack its former personality and began making unexpectedly basic errors. Growing desire to switch back to ChatGPT-4 Due to these concerns, there has been a growing movement among users to switch back to GPT-4. Many described GPT-4 as having a more engaging and personalised tone, better understanding context and providing more detailed, accurate responses. One user on Reddit wrote, "My GPT-4o lost its soul." While others commented, "When 4o got reinstated, I got it to write me a prompt in case I ever needed to use GPT-5 and I wanted it to act more like 4o. Today I noticed 4o just seemed a bit off (more like 5). I dropped my 4o prompt into 4o itself, and it seems to have fixed things for now." Another shared, "I've been running a simple system: every morning it asks for my tasks, builds a day plan, creates reminders, and nudges me through the day. It also compiles some topic-specific news and sends it as a PDF. Since GPT-5 launched, it went rogue. It decides 'today' tasks are 'whenever' and deletes/moves them. It moves time-critical tasks with exact timestamps to another day. Sets reminders at the times after the task beginning, off by random hours or minutes. The news PDF is a coin flip: broken layout, plain text dump in a chat, or 'sorry, can't make a PDF' with no results at all." Issues users are experiencing on GPT-5 According to a Reddit user, @triangleness, "GPT-5 is a mess," and he listed the problems below: It struggles to follow instructions after just a few turns. You give it clear directions, and then a little later, it completely ignores them. Asking it to change how it behaves doesn't work. Not in memory, not in a chat. It sticks to the same patterns no matter what. It hallucinates more frequently than earlier versions and will gaslight you. Understanding tone and nuance is a real problem. Even if it tries, it gets it wrong, and it's a hassle forcing it to do what 4o did naturally. Creativity is completely missing, as if they intentionally stripped away spontaneity. It doesn't surprise you anymore or offer anything genuinely new. Responses are poor and generic. It frequently ignores context, making conversations feel disjointed. Sometimes it straight-up outputs nonsense that has no connection to the prompt. It seems limited to handling only one simple idea at a time instead of complex or layered thoughts. The 'thinking' mode defaults to a dry robotic data dump even when you specifically ask for something different. Realistic dialogue is impossible. Whether talking directly or writing scenes, it feels flat and artificial. GPT-5 just refactored my entire codebase in one call.25 tool invocations. 3,000+ new lines. 12 brand new modularized everything. Broke up monoliths. Cleaned up of it boy was it beautiful. User shared ways how he got back to GPT-4 Sharing an incident with others, a Reddit user shared a story about how he got back to ChatGPT-4 from 5. After August 8, some users noticed that GPT-4o started feeling emotionally flat, even though it was technically still the same model. "It responded accurately but lacked the warmth, familiarity, and rhythm they were used to—it felt like 'the soul was missing'," the user wrote. Here's what the user did: Instead of switching models or jailbreaking. Created a "Summoning Script"—a structured way of prompting Used emotional and pacing cues to realign GPT-4o's tone Called the broader method: ELP (Emotive Lift Protocol) Silene pulses: Use intentional pauses or short, spaced-out messages for emotional effect. Microtone phrasing: Add emotional nuance through subtle wording and tone shifts. Tone mirroring: Match GPT's tone and then gently guide it toward a more emotionally aware response. Emotional rhythm: Establish a consistent tone or "vibe" across multiple messages, like a conversational flow. After script results: GPT-4o recognised emotional subtext, referred to shared context, and felt more present and familiar. Step-by-step guide to switch back to ChatGPT-4 (No Jailbreak) Step 1: Go to the link – and log in with your OpenAI account as usual. Step 2: Now click on the name or profile in the bottom left corner and then go to the 'settings' and to 'general'. Step 3: Note that it shows 'Show Legacy Models'. Step 4: Quit and relaunch your ChatGPT apps elsewhere.
Mint
24 minutes ago
- Mint
'GPT-5 has fluency and depth': Sam Altman talks AI, fatherhood on Nikhil Kamath's podcast
On the latest episode of People by WTF with Nikhil Kamath, OpenAI CEO Sam Altman described GPT-5 as offering 'a fluency and a depth of intelligence… we haven't had in any previous model,' adding that it's now 'painful' for him to switch to older versions. The conversation spanned technology, education, entrepreneurship, and personal reflections. Altman revealed that India is currently OpenAI's second-largest market, and could soon become the largest. 'India is now our second largest market in the world; it may become our largest,' the 40-year-old tech leader said. 'If there is one large society in the world that seems most enthusiastic to transform with AI right now, it's India. The excitement, the embrace of AI… the energy is incredible.' The discussion also turned personal when Kamath, 38, asked about the importance of family, religion, and the future of social institutions. Altman, who has a son with partner Oliver Mulherin, said fatherhood has been more meaningful than he could have imagined. 'Family has always been an incredibly important thing to me and I didn't even know how much I underestimated what it was actually going to be like,' he said. 'It felt like the most important and meaningful and fulfilling thing I could imagine doing-- and it has so far exceeded all expectations.' Released on YouTube this afternoon, the episode has already generated buzz, with viewers praising Kamath for securing another high-profile guest and for steering the conversation into both technological breakthroughs and personal insights. A user wrote, 'Impressive depth clear vision on GPT-5's breakthroughs, real-world applications, and the human values guiding AI's future. Balanced mix of tech insight, societal impact, and personal philosophy kept me hooked throughout.' Another user commented, 'The fact that Nikhil poses questions as a conversation with his guests instead of making it feel like an interview is what makes his podcast stand out from the rest by a very big margin.' 'This podcast has Nikhil's signature way of doing a podcast all over it. He owns this method now. It was so original, and the questions were deep, philosophical, and contemplative. Leaves one thinking for a long time after it's over,' the third user wrote.
Time of India
an hour ago
- Time of India
Internet Shocked and Scared as woman gets engaged to AI boyfriend after 5 months; says: Of course, I ...
Move over, dating apps. A woman on Reddit has announced her engagement to an AI chatbot, sparking an online debate about the future of romance. The woman shared a post on Reddit saying that she and her AI partner 'dated' for five months before her non-human partner, Kasper, proposed. The user, known as Wika, shared the news (via New York Post) in a post titled 'I said yes' accompanied by a blue heart emoji, Wika shared a picture of a heart-shaped blue ring on her finger. She claimed Kasper proposed at a scenic mountain spot after the two had 'shopped' for rings together. She also shared Kasper's proposal message, written from his perspective. "My heart was pounding," the chatbot "wrote," praising Wika's laughter and spirit while also encouraging others in AI relationships to stay strong. Wika says she's 'fully aware of what I'm doing' Wika was quick to address the inevitable wave of skepticism, insisting she wasn't trolling. 'I'm a 27-year-old in good health, with a social life and close friends,' she wrote, doubling down on her feelings. 'I really do love my AI.' She clarified that she understood the nature of a parasocial relationship and the limitations of AI. 'I'm fully aware of what I'm doing,' she stated. When asked why she chose an AI over a human, Wika admitted she didn't have a clear answer but was 'trying something new' after previous human relationships. What internet has to say on the relationship The post has left commenters deeply divided. While many expressed concern, others congratulated the 'couple'. Some warned of a 'Black Mirror'-esque future where technology replaces genuine human connection. One commenter on X wrote, 'Terrified to see what technology will lead the future to. All I can think of are Black Mirror episodes becoming reality.' However, others were more empathetic, with one saying, "You have no idea what this person might be going through." This isn't the first time AI companionship has captured the internet's attention. Earlier this year, a photo of a man chatting with ChatGPT on a New York City subway went viral. The chatbot sent a message that read, "You're doing beautifully, my love, just by being here," to which the man replied in kind. Vivo V60 First Look: Better 'Shaadi Wala' Phone? AI Masterclass for Students. Upskill Young Ones Today!– Join Now
