GPT-5 preview: 5 must-know facts about the ChatGPT upgrade even Sam Altman fears

India Today

a day ago

As speculation builds around OpenAI's next big AI model, GPT-5, the conversation has taken an unexpected turn, which is led by OpenAI's own CEO, Sam Altman. From cryptic social media posts to comparisons with historic weapons projects, Altman's recent remarks have made one thing clear: GPT-5 may be more powerful than even he is fully comfortable with. Here are five key things to know about the upcoming upgrade:advertisement-Sam Altman says GPT-5 made him 'nervous'During a podcast with comedian Theo Von, Altman admitted to feeling 'very nervous' while testing GPT-5, calling it 'very fast.' But it was his comparison to the Manhattan Project, which was a secret World War II programme that developed nuclear weapons, that really captured attention. 'There are no adults in the room,' he warned, criticising the lack of regulation around advanced AI tools.Altman added that the growing reliance on AI for everyday decisions 'feels bad and dangerous,' raising concerns that society may be sleepwalking into a future where AI holds too much influence.-A cryptic screenshot sparks online debate
The internet got its first possible glimpse of GPT-5 through an offhand interaction on X (formerly Twitter). When Altman praised the animated series Pantheon, a user asked if GPT-5 recommended it. In response, Altman posted a screenshot showing a chatbot praising the show and citing critic scores. While there was no confirmation, many believed it was a subtle preview of GPT-5's new abilities.The tone and structure of the chatbot's reply echoed older ChatGPT versions but hinted at more precise, summarised outputs, fueling speculation about search-like upgrades.
-Grok jumps in to add to the hypeElon Musk's AI chatbot Grok joined the online conversation, acknowledging the growing expectations around next-gen models. 'Coherent answers are a win,' Grok posted, while teasing its own upcoming features and asking users what they would 'borrow' first.The exchange highlighted the intense competition in the AI space, with platforms like Grok, Gemini, Claude and Meta's Llama all pushing to outdo one another.-Users mock the hype, lightlyNot everyone was swept away by the speculation. Some users joked that GPT-5 was being treated like a miracle worker. One person remarked that expectations were so high people thought it should 'leap out of the screen and 3D print answers.' Another said just getting a stable response should count as a success.The reactions underline a growing tension between marketing buzz and user expectations.-Bigger questions around safety and control remainAltman's comments bring the focus back to responsibility. He has long warned of AI's risks, but his latest remarks felt more personal. With GPT-5 on the horizon, he seems to be grappling not just with its capabilities, but with what its power could mean for society.His honesty may be rare in a space dominated by hype, but it could also be a warning sign the world needs to take seriously. Or, could it just be a very unique marketing strategy?- EndsMust Watch