An internal memo obtained by sources reveals that OpenAI CEO Sam Altman declared a "Code Red" status on Monday, directing full resources toward enhancing ChatGPT's quality while postponing other product developments.

Altman emphasized that OpenAI must significantly improve the chatbot's daily usability, including refining personalization features, boosting speed and reliability, and expanding its knowledge base.

To accelerate progress, OpenAI will delay initiatives such as its advertising business, AI agents for healthcare and shopping, and a personal assistant project called Pulse. Employees were encouraged to temporarily reassign teams, with daily stand-up meetings scheduled for ChatGPT-focused groups.

Nick Turley, OpenAI's ChatGPT lead, later confirmed on X that the company is prioritizing user growth while making the experience "more intuitive and personalized."

The memo signals OpenAI's mounting pressure from rivals rapidly closing its AI lead. Google's newly launched Gemini AI model recently outperformed OpenAI in benchmark tests, sending Alphabet (GOOGL) shares higher. Since August, Gemini's user base has surged from 450 million to 650 million monthly active users. Anthropic also poses growing competition for enterprise clients.

Despite pledging billions for data centers, OpenAI's path to monetizing these investments remains uncertain. Though still private—with CFO Sarah Friar stating no IPO plans in November—its fortunes are intertwined with tech giants like NVIDIA (NVDA), Microsoft (MSFT), and Oracle (ORCL).

Unlike cash-rich competitors, OpenAI relies on continuous fundraising. Its financial projections indicate expenses far exceeding Anthropic's, requiring $200 billion in revenue by 2030 to achieve profitability.

Altman has largely assuaged financial concerns by highlighting ChatGPT's 800 million weekly active users and OpenAI's research leadership. He noted an upcoming reasoning model would outperform Google's Gemini, with other advancements underway.

Recent months saw OpenAI struggling to balance ChatGPT's safety and engagement. Its August GPT-5 release drew criticism for overly rigid responses and inaccuracies in basic queries, prompting a tone and accuracy upgrade last month.

The memo referenced a prior "Code Orange" status for ChatGPT improvements, with yellow, orange, and red codes indicating escalating urgency internally.