OpenAI has rolled out its fifth-generation large language model, GPT-5, replacing GPT-4o across ChatGPT and the company’s API on 16 August. The system comes in four sizes—v, mini, nano and chat—and promises stronger reasoning, coding and agent-building abilities. At launch, GPT-4o was immediately removed from the free tier and put behind the existing ChatGPT Plus paywall, while a new cut-price ChatGPT Go plan was introduced at US$5 a month. The abrupt switch triggered a wave of user complaints about higher costs and what many described as a colder, less intuitive conversational style. In response, OpenAI said it will restore access to GPT-4o for paying customers and deploy a software update ‘within days’ to make GPT-5 warmer without reviving the flattery that plagued earlier models. Chief Executive Officer Sam Altman also told reporters the company is grappling with a shortage of graphics chips—“we’re out of GPUs”—forcing it to ration the most capable versions of the model and accept that chatbots “are not going to get much better” until more capacity comes online. Despite the consumer backlash, early enterprise uptake has been brisk. Internal usage figures cited by partners and first reported by CNBC show GPT-5 has already more than doubled coding and agent-building activity and driven an eight-fold jump in complex reasoning workloads. Companies such as Box, Vercel and Cursor have made the model their default, lured by lower prices relative to rival systems. OpenAI has assembled a 500-person sales force to court corporate buyers, while tailoring the product for high-growth markets such as India by adding multilingual support and lower-cost tiers. Altman has used the launch to reiterate his views on the labour market, predicting that artificial intelligence will create better-paid roles for younger workers but warning that employees nearing retirement risk being left behind if they do not reskill. The comments underscore the broader stakes of GPT-5’s debut as OpenAI balances technical progress, economic constraints and mounting social scrutiny.
🤖En una entrevista reciente, el CEO de OpenAI, Sam Altman, advirtió que los trabajadores de mayor edad serán los más expuestos a los cambios provocados por la automatización. | Más información en https://t.co/cAqVjFox4D https://t.co/lDEqw3AyTc
ICYMI: OpenAI CEO Sam Altman is more concerned about AI's effects on soon-to-be retirees than he is about recent graduates. https://t.co/ABFnH8m191
La inteligencia artificial exige replantearse las ventajas e inconvenientes de la utilidad tecnológica y el riesgo. Las respuestas no guiadas de los chatbots, por ejemplo, no pueden limitarse de forma clara. #Facebook #IA Jonathan Guilford https://t.co/tNWM5qxFvI