Latest news with #GPT-4o
Hans India
44 minutes ago
- Business
- Hans India
Sam Altman Raises Alarm Over ‘Self-Destructive' AI Use After GPT-5 Backlash
OpenAI CEO Sam Altman has voiced deep concerns about how some ChatGPT users are interacting with artificial intelligence, warning that a segment of the community may be engaging in 'self-destructive' ways. His comments follow mounting criticism over the company's decision to discontinue popular older AI models, including GPT-4o, as part of the GPT-5 rollout. In a candid post on X, Altman reflected on the surprising emotional attachment users have formed with specific AI models. 'If you have been following the GPT-5 rollout, one thing you might be noticing is how much of an attachment some people have to specific AI models. It feels different and stronger than the kinds of attachment people have had to previous kinds of technology (and so suddenly deprecating old models that users depended on in their workflows was a mistake),' he wrote. Altman emphasized that while technology—including AI—can be a powerful tool for positive engagement, it can also become harmful under certain conditions. 'People have used technology including AI in self-destructive ways; if a user is in a mentally fragile state and prone to delusion, we do not want the AI to reinforce that. Most users can keep a clear line between reality and fiction or role-play, but a small percentage cannot. We value user freedom as a core principle, but we also feel responsible in how we introduce new technology with new risks.' While many people have turned to ChatGPT as a virtual therapist, mentor, or life coach, Altman clarified that such uses were not inherently troubling. 'This can be really good! A lot of people are getting value from it already today,' he noted. His primary concern lies in scenarios where AI guidance might subtly steer users away from choices that support their long-term well-being. According to Altman, the level of trust some users place in ChatGPT for crucial life decisions is both remarkable and worrisome. 'People really trust the advice coming from ChatGPT for the most important decisions,' he said, adding that this trust makes him uneasy. The GPT-5 Backlash The controversy stems from OpenAI's decision to retire older GPT and reasoning models, a move that sparked an outcry on social media. Many long-time users claimed that GPT-5's responses felt shorter, less nuanced, and lacking the emotional depth they had come to rely on. The abrupt transition left some feeling that a key part of their workflow—or even their emotional support system—had been taken away without adequate notice. Faced with this backlash, OpenAI reversed course on some decisions, working to restore certain capabilities and give users more flexibility. However, Altman's remarks underscore a broader challenge for AI developers: balancing innovation and safety while addressing the emotional bonds users form with these tools. As AI becomes more integrated into daily life, Altman's warning serves as a reminder that the technology's impact extends far beyond productivity—touching on mental health, trust, and the boundaries between human and machine relationships.
&w=3840&q=100)
Business Standard
44 minutes ago
- Business Standard
Apple Intelligence on iPhone to get OpenAI's GPT-5 model access with iOS 26
Apple Intelligence on iPhones, iPads, and Macs will soon integrate OpenAI's latest GPT-5 AI model. In a statement to 9To5Mac, Apple confirmed that GPT-5 will be coming to Apple Intelligence with iOS 26, iPadOS 26, and macOS Tahoe 26. At present, Apple Intelligence relies on OpenAI's GPT-4o model, according to the report. Apple's newest operating systems — including iOS 26 for iPhones — are already available to select users through public and developer betas for testing. However, GPT-5 integration is not yet active within Apple Intelligence. It is expected that the AI suite will gain access to OpenAI's latest model with the stable release scheduled for later this year. Apple Intelligence currently leverages ChatGPT for several functions: Use Siri to access ChatGPT: Siri can draw on ChatGPT for certain requests, including answering questions about photos and documents. Use ChatGPT with Writing Tools: ChatGPT can generate text or images from a simple description. Use ChatGPT with visual intelligence: Through Camera Control, visual intelligence can help quickly identify and learn more about surrounding places and objects. These capabilities currently run on the GPT-4o model and are expected to transition fully to GPT-5 once iOS 26, iPadOS 26, and macOS Tahoe 26 roll out. Apple is also set to bring ChatGPT-powered image generation to its Image Playground app with the upcoming software releases. OpenAI GPT-5: What is new OpenAI has described GPT-5 as its most advanced 'AI system' yet, delivering state-of-the-art performance in coding, mathematics, writing, health, visual recognition, and more. The model is designed to better interpret context and adjust its response style accordingly — delivering quick answers when possible or taking extra time for more detailed, thoughtful replies when needed. This unified architecture incorporates a deeper reasoning engine — known as GPT-5 Thinking — along with a real-time router that determines whether to give a fast or in-depth response based on the flow of the conversation. According to OpenAI, GPT-5 not only surpasses previous versions in benchmark tests but also performs better with real-world prompts. It is said to follow instructions more accurately and reduce hallucinations compared to earlier models.
&w=3840&q=100)
Business Standard
an hour ago
- Business Standard
OpenAI restores GPT-4o to ChatGPT Plus after user backlash over removal
Sam Altman confirmed GPT-4o will return to ChatGPT Plus after its removal during GPT-5 rollout sparked backlash from users OpenAI chief Sam Altman has confirmed that GPT-4o will return to ChatGPT Plus shortly after its removal during the GPT-5 rollout. The decision follows widespread user backlash, with many expressing emotional attachments to the model. 'We for sure underestimated how much some of the things that people like in GPT-4o matter to them, even if GPT-5 performs better in most ways,' Altman posted on X (formerly Twitter). He later added, 'We will let Plus users choose to continue to use 4o.' Why GPT-4o was removed OpenAI launched GPT-5, GPT-5 Thinking, and GPT-5 Thinking Pro on August 7, replacing all earlier ChatGPT models, including GPT-4o. Many users lamented the change, with one describing the loss as 'like killing a friend of millions of users.' Models replaced by GPT-5 included: GPT-4o GPT-4.1 GPT-4.5 GPT-4.1-mini o4-mini o4-mini-high o3 o3-pro When older conversations are reopened, ChatGPT now switches them to the nearest GPT-5 equivalent — for example, chats using GPT-4o or GPT-4.1 default to GPT-5, except for Plus users. What's new in GPT-5 OpenAI describes GPT-5 as its most advanced AI system, excelling in coding, mathematics, writing, healthcare, and visual understanding. It adapts its pace depending on context — delivering quick replies when speed is essential and detailed output when depth is needed. Key updates include: Unified system: GPT-5 merges previous versions into one model with an upgraded reasoning engine, GPT-5 Thinking. Improved performance: Notable gains in writing, coding, and health queries, with fewer inaccuracies. Personalities: Four new optional personalities — Cynic, Robot, Listener, and Nerd. Voice mode: Now works with custom GPTs, with lifted or expanded limits for paid and free users, plus dynamic speaking instructions. Why GPT-4o is coming back Despite GPT-5's improvements, many users said GPT-4o felt more relatable and consistent in conversations. Altman's decision to restore it for Plus subscribers acknowledges this emotional bond, giving users the option to keep using a model they trust. First Published: Aug 11 2025 | 10:55 AM IST
Mint
an hour ago
- Business
- Mint
Why ChatGPT Plus users think GPT-5 is a downgrade — and how OpenAI is trying to win them back
Ever since the rollout of OpenAI's GPT-5 model last week, there has been growing disquiet among the company's users over its performance compared to previous models. GPT-5's tendency to answer questions briefly and in an emotionally distant manner has led many users to label it as a cost-saving exercise by the company. What didn't help was that, as GPT-5 started rolling out to all users, OpenAI also removed access to older models including GPT-4o, o3, o3 Pro, o4-mini, and more. While ChatGPT Pro and Team subscribers could still access these models via the legacy settings, they were no longer available to Free or Plus tier subscribers. While OpenAI kept the rate limits for GPT-5 on ChatGPT Plus more or less the same as GPT-4o, many still felt it was a downgrade compared to the earlier situation. Let's take a look at why — and what OpenAI is doing to address these concerns. With GPT-5, ChatGPT Plus users get access to 80 messages every 3 hours for the standard model and 200 messages per week for the GPT-5 Thinking model. Prior to the new model launch, however, Plus subscribers had access to multiple models with much higher rate limits: o3: 100 messages per week o4-mini high: 700 messages per week o4-mini: 2,100 messages per week GPT-4o: 80 messages per 3 hours Naturally, Plus subscribers felt the new launch resulted in a downgrade, given the significant reduction in the number of messages they could send. Moreover, OpenAI did not increase the context window for Plus users, which remained at 32,000 tokens (around 24,000 words) compared to the 128,000-token (around 96,000 words) context window available on higher tiers. A longer context window allows an AI system to remember and process much longer texts in a single conversation. The GPT-5 launch triggered a wave of criticism on social media, particularly on ChatGPT and OpenAI subreddits, where many users threatened to cancel their Plus subscriptions. This led to immediate action from OpenAI. As an initial step, OpenAI doubled the rate limit for GPT-5 on the Plus plan to 160 messages per 3 hours and enabled access to GPT-4o for these users via the same legacy model settings available to Pro subscribers. On Monday, Sam Altman also promised to raise the limits for the GPT-5 Thinking model, stating that the company is aiming to provide 1,000 messages per week.
Hans India
2 hours ago
- Hans India
GPT-5 Rollout Sparks Backlash from ChatGPT Plus Users — OpenAI Promises Higher Limits
OpenAI's latest model, GPT-5, has barely been out a week, yet it's already stirred up a storm among ChatGPT Plus subscribers. While the company billed GPT-5 as a smarter, more capable AI, many users feel it's a step back from previous offerings — both in personality and in how much they can use it. One of the most common complaints? GPT-5's tone. Users say the new model's answers are shorter, less nuanced, and come across as 'emotionally distant.' For many, it feels like a cost-cutting shift rather than a meaningful upgrade. The criticism intensified when OpenAI quietly removed access to older models such as GPT-4o, o3, o3 Pro, and o4-mini for Free and Plus subscribers. While Pro and Team plan users could still reach these models through legacy settings, Plus users — who make up a large part of ChatGPT's paying base — were left without the variety they had grown used to. Adding to the frustration were the new message limits. Under GPT-5, Plus users receive 80 messages every three hours for the standard model and just 200 per week for the GPT-5 Thinking model. By comparison, before GPT-5's release, subscribers enjoyed far more flexibility: o3: 100 messages per week 100 messages per week o4-mini high: 700 messages per week 700 messages per week o4-mini: 2,100 messages per week 2,100 messages per week GPT-4o: 80 messages per three hours On top of that, the context window — which determines how much past conversation the model can 'remember' — remains at 32,000 tokens (around 24,000 words) for Plus users, far smaller than the 128,000-token window offered to higher-tier subscribers. For users handling large or complex projects, that's a meaningful limitation. Unsurprisingly, online forums such as the ChatGPT and OpenAI subreddits lit up with complaints, and some users openly threatened to cancel their subscriptions. The backlash was loud enough that OpenAI moved quickly to respond. As a first step, the company doubled the GPT-5 rate limit for Plus users to 160 messages per three hours. It also restored access to GPT-4o through the same legacy settings that Pro subscribers use. But perhaps the most attention-grabbing promise came from OpenAI CEO Sam Altman, who posted on Monday that the company is working to raise the GPT-5 Thinking model limit to 1,000 messages per week. Whether these changes will be enough to win back disillusioned subscribers remains to be seen. But one thing is clear: GPT-5's launch has been a reminder that for AI companies, technical upgrades aren't the only measure of success — user trust and experience matter just as much.
