OpenAI Shockingly Releases GPT-5: The World's First Super AI Model Opening Reasoning Function to Free Users Officially Launched

News Summary

OpenAI officially launched GPT-5, its most advanced artificial intelligence model to date, to all users worldwide on August 7. The model demonstrates significant improvements in areas such as reasoning capabilities, programming, writing, and healthcare, and for the first time, opens up inference functionality to free users, marking a new milestone in AI technology.

Breakthrough Launch Shakes the AI World

OpenAI officially released GPT-5, its latest and most advanced large artificial intelligence model, on August 7. This launch is regarded by the industry as a significant milestone in the history of AI development. Sam Altman, OpenAI's CEO, stated in a briefing with reporters, "I tried going back to GPT-4, and it felt pretty bad," vividly illustrating the new model's immense progress.

Historic Moment: Full User Coverage

GPT-5 has been fully rolled out to OpenAI's free users, Plus users, Pro users, and Team users. This marks the first time in OpenAI's history that an inference model—a type of model capable of "thinking" or performing an internal chain of thought before responding—has been made available to free users. This move breaks down the paywall for advanced AI functionalities, allowing a wider user base to experience cutting-edge artificial intelligence technology.

Comprehensive Leap in Technical Performance

GPT-5 has achieved significant breakthroughs across multiple key metrics. The model scored 94.6% on the AIME 2025 for mathematics, obtained a 74.9% SWE-bench Verified score in practical programming tasks, and achieved an 84.2% MMMU score for multimodal understanding. Even more remarkably, GPT-5's hallucination rate has been significantly reduced, with its responses being 45% less likely to contain factual errors than GPT-4o.

Significant Enhancement in Safety and Reliability

OpenAI has invested unprecedented effort in the safety development of GPT-5. The company conducted extensive safety evaluations, including 5,000 hours of testing. GPT-5 employs a "safety completion" mechanism, which does not completely refuse to answer potentially risky questions but instead provides high-level responses within safety constraints.

Michelle Pokrass, OpenAI's Post-Training Lead, stated, "GPT-5 is trained to recognize when it can't complete a task, avoid speculation, and explain limitations more clearly, which reduces unfounded claims compared to previous models."

Revolutionary Programming Capability Demonstration

During the launch demonstration, OpenAI showcased GPT-5's "vibe programming" capability. The company asked GPT-5 to create a web application to help English speakers learn French, requiring an engaging theme, activities like flashcards and quizzes, and a way to track daily progress. Amazingly, GPT-5 generated two different applications in a matter of seconds.

Positive Feedback from Enterprise Users

Enterprise users have shown great enthusiasm for GPT-5. Aaron Levie, CEO of Box, stated that previous AI models failed in the company's most advanced tests because they struggled to understand complex mathematics or logic in long documents, but GPT-5 is a "complete breakthrough." He noted, "The model is able to retain more information it's looking at and then use higher-level reasoning and logic capabilities to make decisions."

Market Impact and User Scale

OpenAI anticipates ChatGPT's weekly active users to reach 700 million this week, demonstrating its massive user base. The company is currently in talks with investors for an equity sale at a valuation of approximately $500 billion, a valuation that reflects the market's high recognition of OpenAI's technological prowess.

Multi-Version Deployment Strategy

OpenAI has released three different versions of the model through its API: gpt-5, gpt-5-mini, and gpt-5-nano, which are optimized for different cost and latency requirements. This tiered strategy ensures that businesses and developers of varying scales can find solutions tailored to their needs.

Deep Integration Partnership with Microsoft

Microsoft CEO Satya Nadella posted on X, stating, "It's hard to believe it's only been two and a half years since Sam Altman met with us in Redmond, Washington, and first showed the world GPT-4." According to a Microsoft blog post, Microsoft 365 Copilot, the consumer-facing Copilot, and Azure AI Foundry, which developers can use to integrate AI models into third-party applications, will all quickly gain access to GPT-5.

Future Outlook and Industry Impact

The release of GPT-5 not only represents a new height in OpenAI's technological prowess but also signifies that artificial intelligence has entered a new stage of development. Altman stated that GPT-5 is like having a team of PhD-level experts available at any time, and he believes, "People are limited by ideas, but in many new ways, not really limited by execution capability."

This launch is expected to further accelerate the application of AI technology across various industries, with profound impacts anticipated in fields such as education, healthcare, programming, and creative industries. As free users gain access to advanced AI functionalities, the democratization of artificial intelligence will significantly accelerate, opening the door to an intelligent future for hundreds of millions of users worldwide.