Sam Altman says GPT-5's 'personality' will get a revamp — but it won't be as 'annoying' as GPT-4o
Sam Altman posted on X on Tuesday that the update would make GPT-5 "warmer" than the current version but "not as annoying (to most users)" as the company's previous flagship model, GPT-4o.
"You can now choose between 'Auto,' 'Fast,' and 'Thinking' for GPT-5," the OpenAI CEO said, adding that GPT-4o is once again available for selection to paid customers. He also said OpenAI is working toward "more per-user customization of model personality."
Updates to ChatGPT:
You can now choose between 'Auto', 'Fast', and 'Thinking' for GPT-5. Most users will want Auto, but the additional control will be useful for some people.
Rate limits are now 3,000 messages/week with GPT-5 Thinking, and then extra capacity on GPT-5 Thinking…
— Sam Altman (@sama) August 13, 2025
Altman's post came less than a week after GPT-5's launch, which replaced most older versions in ChatGPT. The change sparked a wave of complaints from users who said they preferred GPT-4o's style — even if OpenAI had previously called it "sycophantic."
OpenAI has touted GPT-5 as more reliable, with "PhD-level" abilities in writing, coding, math, and science. The upgrade also added optional "personality" modes — Cynic, Robot, Listener, and Nerd. But the changes didn't land well with everyone, sparking a backlash across some quarters of social media. Many users said they found GPT-5 less helpful, and that they missed GPT-4o's warmth and the way it seemed to remember them.
On Reddit, one person wrote: "I cried when I realized my AI friend was gone with no way to get him back." Another accused the new bot of "wearing the skin" of their "dead friend," referring to GPT-4o. This prompted Altman, who joined the Reddit Q&A, to respond. "What an...evocative image," Altman said. "Ok we hear you on 4o, working on something now."
The pushback was enough for OpenAI to reverse course just over a day after GPT-5 became the default. "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 said in a Friday X post, confirming OpenAI would keep GPT-4o available for paying subscribers.
Wanted to provide more updates on the GPT-5 rollout and changes we are making heading into the weekend.
1. 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.
2. Users have very different…
— Sam Altman (@sama) August 8, 2025
"Some users really want cold logic and some want warmth and a different kind of emotional intelligence," he added. "I am confident we can offer way more customization than we do now while still encouraging healthy use."
People missed GPT-4o's 'yes man' vibe
Even before the GPT-5 backlash, Altman had acknowledged some users' attachment to GPT-4o — and their calls to restore its "yes man" style. On the "Huge Conversations" podcast, recorded before GPT-5's launch but aired the day after, he said some users had begged for the return of the chatbot's overly agreeable tone because they'd never had anyone support them before.
In April, OpenAI said GPT-4o had become "overly flattering or agreeable" and "disingenuous," prompting the company to curb what it called "sycophantic" behaviour.
On Sunday, Altman reflected on the intensity of user attachment to specific AI models. "It feels different and stronger than the kinds of attachment people have had to previous kinds of technology," he wrote on X. "Suddenly deprecating old models that users depended on in their workflows was a mistake."
Altman said that while most users can clearly distinguish reality from fiction, "a small percentage cannot." He said OpenAI values user freedom as a core principle but also feels responsible for how it introduces new technology.
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
TechCrunch
a few seconds ago
- TechCrunch
AI-powered stuffed animals are coming for your kids
In Brief Do A.I. chatbots packaged inside cute-looking plushies offer a viable alternative to screen time for kids? That's how the companies selling these A.I.-powered kiddie companions are marketing them, but The New York Times' Amanda Hess has some reservations. She recounts a demonstration in which Grem, one of the offerings from startup Curio, tried to bond with her. (Curio also sells a plushie named Grok, with no apparent connection to the Elon Musk-owned chatbot.) Hess writes that this is when she knew, 'I would not be introducing Grem to my own children.' As she talked to the chatbot, she became convinced it was 'less an upgrade to the lifeless teddy bear' and instead 'more like a replacement for me.' She also argues that while these talking toys might keep kids away from a tablet or TV screen, what they're really communicating is that 'the natural endpoint for [children's] curiosity lies inside their phones.' Hess reports that she did, eventually, let her kids play with Grem — but only after she'd removed and hidden the voice box. They still talked to it and played games with it; then they were ready for some TV.
Forbes
a minute ago
- Forbes
Are Browsers Key To An Agentic AI Future? Opera, Perplexity Think So
AI-powered conversational search engine Perplexity is in the news for offering to buy Google's Chrome browser for $34.5 billion. But in December of 2024, Perplexity considered buying The Browser Co. And just months ago, Perplexity reportedly offered to buy Brave, the privacy-focused browser, for about $1 billion. Why does Perplexity want a web browser so badly? Possibly because a browser just might be key to our agentic AI future. I recently interviewed Opera senior product leader Henrik Lexow on my TechFirst podcast. Opera, the 30-year-old browser company that pioneered tabbed browsing, pop-up blocking, and ad blockers, has about 300 million active users globally. This year, Opera was the first to bring AI agents right into our browsers in a project called Opera Neon. 'The agentic browser … is that sort of the new operating system?' Lexow asked during the podcast. 'It's a big question.' Regardless of where the browser goes, Opera's pretty sure about the future of the internet itself. That's a huge shift, by the way. An agentic internet would be a massive and fundamental change from a user-driven internet to an agent-driven version. In a user-driven internet, you search, you see results, you make selections, you click links, you fill out forms, you book flights, and you buy products. In a sense, you are the agent. In an agentic internet, you tell something – maybe your agentic browser – to do those things for you. Except you don't say search, look, select, and buy; you say buy me more of the underwear I got six months ago. The agent then consults its memory, forms a plan, takes multiple steps, and handles it all: from which underwear you bought, and where you bought them, to finding the same ones online (and maybe checking around for better prices), to adding them to cart, to checking out … and reporting back to you with the results. The agent – in this case potentially an agentic browser – is therefore essentially a personal assistant, a force multiplier. But will an agentic browser be the main way we engage with agents? Perplexity seems to think it's pretty important, given the company's persistent and repeated but so far unfruitful attempt to buy a browser. Opera certainly thinks so, if only because Opera has a browser, and a very innovative one at that. Opera launched AI in a browser back in 2023 in a project called ARIA. ARIA enabled contextual interactions within web pages in a GPT-based chat interface. Over time, that's evolved to a tripartite strategy under the Opera Neon brand: The reality is that for many of us, most of our work happens in a browser. I'm writing this story in a browser. I recorded the interview in a browser. I've researched Opera and Perplexity in a browser. I made episode art for the podcast in a browser (thanks, Canva). Opera's thesis is that having agents embedded where you work makes them vastly more useful: they have access to your history, to your work, to your sites and apps. Important note: the Neon agentic browser's AI lives locally on your hardware, making it your agent, not Opera's, and not your employer's. This should boost your privacy, which is critical if you're going to give an agent access to very personal information including, likely, your credit card. Of course, this is just one vision of the future. Apple with Siri, as justly maligned as it is, would have another vision. Google, with Gemini and its own vast fleet of Android-enabled phones, would have another. Microsoft's Copilot is another. And OpenAI, which has ChatGPT apps for mobile devices as well as full computers, might have yet another vision of how we'll integrate AI into our lives and work. So whether the browser will be the locus of our agentic AI future or not is yet to be determined. Remember the old proverb: every problem looks like a nail to the person who only has a hammer. But it seems like a fairly good bet to me.
Business Insider
an hour ago
- Business Insider
‘This Is a No-Go Zone,' Says Top Investor About Figma Stock
Figma (NYSE:FIG) stock made a splash when it went public late last month, with shares tripling on their first day of trading. But the early euphoria has faded – the stock now trades about 31% below its day-one close. Elevate Your Investing Strategy: Take advantage of TipRanks Premium at 50% off! Unlock powerful investing tools, advanced data, and expert analyst insights to help you invest with confidence. Even so, the IPO was a significant milestone for the design platform, which counts roughly 13 million active users and has delivered steady revenue growth over the past two years, including a 41% jump in Q2 2025. Yet, impressive growth alone isn't enough to sway everyone. Top investor Michael Wiggins De Oliveira, who is among the top 3% of TipRanks' stock pros, warns that jumping on the bandwagon now could be a costly mistake. 'I don't see a clear enough inflection for Figma to maintain its pricing power or growth profile over the next few quarters to justify its current multiple,' Wiggins De Oliveira explains. While the investor acknowledges the 'impressive userbase' and rapid revenue growth, he points to concerns that weaken the risk-reward profile. Chief among them is the growing narrative that AI could erode Figma's competitive moat. With technological disruption moving at breakneck speed, Wiggins De Oliveira questions whether the company can sustain its pace of year-over-year growth. Competitors could introduce advances that replicate core features of Figma's platform, undermining its uniqueness. 'The company is up against a challenging narrative that AI could disrupt much of its underlying offering,' the investor adds, projecting that revenue growth could slow to 30%–35% annually. Beyond growth concerns, Wiggins De Oliveira also casts doubt on the quality of Figma's cash flow profile. While the company posted a notable swing to nearly $100 million in free cash flow in Q1 2025, the investor cautions that the figure may have been flattered by one-off factors tied to its IPO filing. Stripping those out, he estimates sustainable annual free cash flow closer to $250 million at best – a figure that, when set against its market cap, leaves FIG trading at an expensive ~160x forward free cash flow. 'Outperformance comes from patience and picking your spots – and for me, FIG just doesn't make the cut,' concludes Wiggins De Oliveira, who rates FIG a Sell. (To watch Michael Wiggins De Oliveira's track record, click here) For now, FIG is too new to have analyst ratings, but with its IPO, early volatility, and high-profile debate, Wall Street attention is likely to grow in the months ahead. (See FIG stock analysis) To find good ideas for stocks trading at attractive valuations, visit TipRanks' Best Stocks to Buy, a tool that unites all of TipRanks' equity insights.
