OpenAI Shifts to "Red Code" to Enhance ChatGPT Amidst Rising Competition

Key Takeaways:

• OpenAI declares "red code" to focus on improving ChatGPT.
• Other projects, such as advertising and personal assistants, are postponed to focus on ChatGPT.
• Intense competition from Google (Gemini) and Anthropic is putting pressure on OpenAI.
• OpenAI remains unprofitable and relies on external funding.
• Sam Altman downplays concerns about the company's financial situation.

Full Article:

According to an internal memo viewed by The Wall Street Journal, OpenAI CEO Sam Altman told employees on Monday that the company has declared a "red code" state, fully focusing on improving the quality of ChatGPT, and therefore postponing the development of other product features. This action aims to improve the daily user experience, including better personalization, increased speed and reliability, and expanding the range of answers. Altman indicated that OpenAI will also postpone other projects, including advertising businesses, AI agents for the health and shopping sectors, and a personal assistant called Pulse. He encouraged employees to make temporary team adjustments and indicated that the company will hold daily meetings for the teams responsible for improving ChatGPT. Late Monday, Nick Turley, head of ChatGPT at OpenAI, stated on platform X that the company is currently focused on expanding ChatGPT's user base, while making the experience "more intuitive and personalized." This comprehensive memo is so far the clearest indication that OpenAI is facing tremendous pressure from competitors who are rapidly narrowing the startup's lead in the AI race. What particularly worries Altman is Google, which last month released the next generation of the Gemini AI model that outperformed OpenAI in industry benchmark tests, causing Alphabet (GOOGL.O) shares to rise. Since the launch of the image generation tool Nano Banana in August, the number of Gemini users has continued to climb. Google says its monthly active users have increased from 450 million in July to 650 million in October. At the same time, OpenAI is also facing competitive pressure from Anthropic, which is increasingly popular among corporate clients. OpenAI has pledged to invest hundreds of billions of dollars in building data centers in the future, but the question is when it can translate these investments into substantial revenue, which has caused turmoil in capital markets in recent weeks. Although the company remains unlisted and CFO Sarah Friar indicated in November that there are no plans for an IPO at this time, its fate is closely tied to technology giants such as NVIDIA (NVDA.O), Microsoft (MSFT.O), and Oracle (ORCL.N). OpenAI has not yet made a profit and must continue to raise funds to maintain its operations, which puts it at a financial disadvantage, while technology giants like Google can rely on their revenue to support investments. According to OpenAI's financial forecasts, the company's expenses are much higher than its main competitor Anthropic, and in order to become profitable by 2030, its revenue must grow to approximately $200 billion. To a large extent, Altman has alleviated concerns about OpenAI's financial situation through ChatGPT's huge and steadily growing user base, which currently has more than 800 million weekly active users, as well as the company's leadership in cutting-edge AI research. He said in the memo that the new inference model that OpenAI plans to release next week is superior in performance to Google's latest Gemini model, and that the company is performing well in many other areas. In recent months, OpenAI has worked hard to balance chatbot safety and user engagement. The GPT-5 model released in August had mixed reactions among some users, with critics criticizing it for being too bland and performing poorly in answering simple math and geography questions. Last month, OpenAI upgraded the model to make it more gentle and more capable of accurately understanding and executing user instructions. The memo also mentioned that OpenAI had previously announced an "orange code" state to improve ChatGPT. Insiders revealed that the company internally uses three codes: yellow, orange, and red, to represent varying degrees of emergency.