OpenAI's growth engine is hitting a serious speed bump. The ChatGPT developer recorded just an 18% quarter-on-quarter revenue increase in Q2, a figure that has disappointed some investors and looks especially weak when stacked against rival Anthropic, which not only out-earned OpenAI during the same period but grew at a pace far exceeding market expectations.

According to sources cited by The Wall Street Journal, OpenAI generated $6.7 billion in second-quarter revenue, up from $5.7 billion in Q1. However, operating losses widened further during the period, casting a shadow over the company's much-anticipated IPO prospects.

By contrast, Anthropic's preliminary Q2 revenue surpassed $11.5 billion, representing a quarter-on-quarter jump of more than 140%. This marks the first time Anthropic has overtaken its older rival in quarterly revenue, and the company also managed to post a modest operating profit.

The diverging fortunes of the two firms highlight how dramatically the competitive landscape in AI has shifted this year. With ChatGPT's growth decelerating and Anthropic's coding tool Claude Code enjoying significant success, OpenAI is being forced to rethink its commercial strategy and reshuffle its management team. OpenAI has told investors that growth picked back up in Q3 following the release of a batch of new models in July.

OpenAI's slowing momentum and widening losses

OpenAI's Q2 revenue grew 18% quarter-on-quarter to $6.7 billion, but its operating margin deteriorated further, pushing the goal of profitability ever further out of reach. The performance has disappointed some shareholders who had expected a stronger acceleration from the company.

Meanwhile, OpenAI's management has been through a turbulent stretch recently. Last week, Chief Revenue Officer Denise Dresser stepped down after less than a year in the role. She follows the departures of Chief Operating Officer Brad Lightcap and Fidji Simo, who was once viewed as a potential successor to CEO Sam Altman. The frequent turnover at the executive level has intensified concerns about OpenAI's internal stability.

On the revenue structure front, OpenAI expects enterprise clients to account for more than half of total revenue by the end of this year, signaling an accelerating shift toward the corporate market. According to Axios, co-founder Greg Brockman shared internally last week that the company's latest annualized revenue run rate has reached $40 billion.

Anthropic's rapid rise, overtaking OpenAI in quarterly revenue

Anthropic's performance tells a very different story. The company's preliminary Q2 revenue exceeded $11.5 billion, more than 14 times the figure from the same period last year and up over 140% from Q1's $4.73 billion. This is also the first time Anthropic has surpassed OpenAI in quarterly revenue.

What's more notable is that Anthropic achieved explosive growth while also posting a small operating profit, suggesting its business model is moving toward sustainability. As of the end of July, Anthropic's annualized revenue run rate had surpassed $65 billion, roughly seven times higher than at the end of last year.

One of the key engines behind Anthropic's growth is Claude Code, its developer-focused coding tool. According to PitchBook analyst Harrison Rolfes, while Anthropic's flagship model carries a higher cost per call, its superior accuracy means users don't need to repeatedly run queries or rely on manual review, making the actual total cost per task more competitive.

The IPO race: Anthropic may list first

On the capital markets front, both companies are preparing for IPOs, but Anthropic appears to be moving faster. According to Axios, Anthropic is meeting with potential new investors and plans to go public in September or October this year, with Morgan Stanley, Goldman Sachs, and JPMorgan assisting in the offering.

The strategic importance of listing first should not be underestimated. IPO proceeds will help the company expand the computing power needed for model inference, and being first to the public markets could give Anthropic an extra edge in its rivalry with OpenAI.

On efficiency, Gavin Baker, managing partner at Atreides Management, told Axios that "Anthropic was once far ahead of OpenAI in token efficiency, but OpenAI has narrowed some of that gap." Both companies have struck partnerships with major inference providers and are each developing custom chips to lower the cost of serving AI queries in the future.

It's worth noting that the two companies may use different revenue accounting methods, so direct comparisons should be approached with caution. Additionally, the competitive landscape in AI can shift rapidly, and any single metric's leadership position is unlikely to last. What is clear, however, is that a revenue growth strategy centered on enterprise customers has proven to be the winning formula at this stage.