OpenAI is abandoning consumer products to concentrate resources on building the computational infrastructure and research capabilities it believes will power the next generation of artificial intelligence. In a recent interview, CEO Sam Altman disclosed that the company deliberately shut down Sora, its video generation tool, and Atlas, a web browser he called "the best web browser," to free up resources for what he described as "probably already the most expensive infrastructure project in history".

Why Is OpenAI Killing Successful Products?

The decision reflects a fundamental strategic shift at OpenAI. Rather than competing across multiple product categories, Altman explained that OpenAI is repositioning itself as a platform company, not a product company. The strategy centers on offering a single direct interface and one application programming interface (API), which is a standardized way for software to communicate, to sell artificial intelligence capabilities across the entire cost-performance spectrum.

Altman acknowledged that both products were genuinely good. "Sora was a good product," he stated, and he described Atlas as exceptional. However, the company determined that advancing general intelligence for knowledge work, including building its own computer chips and data centers, took priority over maintaining consumer-facing applications.

Altman

"Most of Altman's effort is on research and compute, which he called probably already the most expensive infrastructure project in history. To free resources, OpenAI killed Sora and its Atlas web browser because the general intelligence for knowledge work mattered more," explained the source material.

Sam Altman, CEO at OpenAI

What Is OpenAI's New Platform Strategy?

OpenAI recently merged ChatGPT and Codex, its coding assistant, into a single interface. Altman's vision is ambitious: he wants 100 million new businesses and 8 billion people building on top of OpenAI's platform. This represents a dramatic expansion from viewing OpenAI as a consumer chatbot company to seeing it as foundational infrastructure for the global economy.

The shift also reflects Altman's candid assessment that artificial intelligence adoption has moved slower than he anticipated. After GPT-4 launched in 2023, he expected rapid disruption of software businesses. Instead, he found that organizational inertia is stronger than expected; companies continue buying from the same vendors and using tools in familiar ways.

How to Understand OpenAI's Long-Term Priorities

• Infrastructure Investment: OpenAI is prioritizing compute and research over consumer products, treating infrastructure as the foundation for all future capabilities.
• Platform Over Products: The company is consolidating interfaces and APIs to serve as a universal platform rather than competing in specific product categories like video generation or web browsing.
• Adoption Challenges: Despite technological breakthroughs, organizational behavior and existing vendor relationships are slowing AI's integration into everyday work, requiring a different approach to market penetration.

When Will AI Have Its "iPhone Moment"?

Altman made a striking admission about the current state of AI adoption: he still uses his computer the old way, scrolling through email and maintaining a to-do list, despite having access to Codex and other advanced tools. He called this "mostly a product failure," noting that artificial intelligence has not yet achieved its "iPhone moment" for how people interface with technology.

He compared the current moment to the smartphone era before Apple's iPhone revolutionized the market. The technological pieces exist, but the product breakthrough that makes AI indispensable to everyday work has not yet arrived. This realization appears to have influenced OpenAI's decision to focus on foundational capabilities rather than consumer applications.

Altman's broader vision extends beyond OpenAI's immediate product roadmap. He predicted "the greatest boom ever in people starting small businesses, empowered by AI," suggesting that the real value of artificial intelligence will emerge through entrepreneurship and platform-based innovation rather than through polished consumer products.

The decision to kill Sora and Atlas signals that OpenAI believes the path to transforming how people work lies not in building the best individual applications, but in creating the most powerful underlying infrastructure. Whether that bet pays off will depend on whether developers and businesses can indeed build the breakthrough applications that AI has not yet produced on its own.