07.08.2025
2 min
555
OpenAI has launched a new version of its chatbot GPT-5, which it calls “smarter, faster, and more useful.” The model is being marketed as a PhD-level expert with fewer errors and more programming capabilities, and it’s now available to all users.
OpenAI CEO Sam Altman said that GPT-5 ushers in a new era in ChatGPT’s development, demonstrating the ability to provide deep expert answers and build software “from scratch.” The model has improved logical thinking skills, answers queries more honestly and accurately, and is less likely to invent information.
OpenAI emphasizes that GPT-5 is capable of working as a full-fledged assistant for developers, demonstrating better coding skills and explaining its decisions. The company has also implemented changes for healthier interactions with users — for example, the AI will not give final advice on sensitive personal issues, but will help weigh arguments.
The previous models, GPT-3 and GPT-4, were considered high school and university-level, respectively. According to Altman, GPT-5 is the first to create the feeling of communicating with an expert in any field. The launch came amid competition from major tech companies, including Anthropic and Elon Musk’s Grok, which also claimed to be “above PhD level.”
At the same time, some ethics experts, including Professor Carissa Velis, emphasize that these achievements may be partly a marketing ploy, and the development of AI needs tighter regulation.
GPT-5 is already available to all users and offers a significantly improved experience with ChatGPT. Whether it really lives up to OpenAI’s ambitious claims will become clear in the coming days, but it is already clear that the model is able to strengthen the company’s position in the global AI race.
