OpenAI's Revenue Soars to $20 Billion, Fueled by a Massive Bet on Computing Power

OpenAI’s financial engine is roaring. The company’s annualized revenue has vaulted from $2 billion in 2023 to more than $20 billion in 2025, according to Chief Financial Officer Sarah Friar. This staggering growth, detailed in recent disclosures, is directly tied to an enormous expansion of the computing infrastructure powering its artificial intelligence systems.

Friar, who joined OpenAI in 2024, noted that the company’s computing capacity grew nearly ninefold over the same period, from 0.2 gigawatts to 1.9 gigawatts. This parallel surge highlights a core strategy: scaling physical hardware to meet explosive demand for products like ChatGPT. Enterprise subscriptions and API usage have been primary drivers of the revenue jump, which increased 233% in 2025 alone.

Yet this ambition carries a staggering price tag. OpenAI is now spending approximately $17 billion more than it earns each year, with the vast majority of that cash consumed by computing costs. This burn rate raises pointed questions about long-term sustainability as the company considers a public offering, potentially in late 2026.

The journey from research project to commercial titan began with ChatGPT’s public release in late 2022. What started as an experiment quickly became a billion-dollar-a-month business by mid-2025. To support this, OpenAI has secured over a trillion dollars in infrastructure commitments with partners like Microsoft, building data centers at a rapid clip.

In response to financial pressures, OpenAI is testing new revenue streams, including a lower-cost subscription tier and introducing advertisements in its free service. The company has also brought on seasoned executives to steer its commercial strategy.

Friar defends the aggressive spending, arguing that revenue growth has kept pace with the investment in computing. As OpenAI eyes a historic IPO and continues its pursuit of more advanced AI, the world is watching to see if this high-stakes bet on scale will ultimately pay off.

Source: Webpronews