OpenAI pivots to revenue focus with former Slack executive named CRO

Shift from tech-first operations to revenue-led management
Persistent bubble concerns amid long-term loss forecasts
IPO plans put the “AI bubble” narrative to the test

OpenAI has moved decisively to recruit external leadership as it transitions from a technology-centric operating model to one centered on revenue generation. With enterprise demand rising rapidly, the move signals an unmistakable acknowledgment that the company can no longer ignore its growing financial pressures. Yet despite these adjustments, market sentiment remains mixed: concerns persist over OpenAI’s structurally high costs and prolonged losses, and its ongoing discussions around an eventual public listing are meeting both anticipation and caution.

Financial burdens expand beneath strong headline performance

On the 9th (local time), OpenAI announced in an official statement that it had appointed Dennis Dresser, former CEO of Slack Technologies, as its Chief Revenue Officer (CRO). On the same day, Pizzi Simo, CEO of OpenAI’s applications division, wrote on social media that “we are at the gateway to placing AI tools in the hands of millions across every industry,” adding that Dresser’s prior experience leading such transformations would be instrumental in turning AI into a useful and trusted resource for enterprises.

The company’s decision to elevate revenue generation reflects its escalating cost structure. According to an internal leak obtained by technology blogger Ed Zitron, Microsoft received approximately 493.8 million USD in revenue-sharing payments from OpenAI last year. By the third quarter of this year, cumulative revenue sharing had climbed to 865.8 million USD. Given that OpenAI pays roughly 20 percent of its revenue to partners, the company’s revenue for the first three quarters of this year is estimated to exceed 4.3 billion USD.

Such figures may appear favorable at first glance. However, behind this growth lies a rapid increase in inference costs that far surpass revenue, creating mounting pressure on cash flow. Inference costs represent the computing expenditures required to process user requests on already-trained models, and they constitute the core expense for enterprise APIs and ChatGPT operations. OpenAI spent roughly 3.8 billion USD on AI model inference in 2024, and in just the first nine months of this year, those expenses reached 8.65 billion USD. The fact that cost growth is outpacing revenue illustrates the financial strain placed on the company as its model operations scale.

Another major concern is the pace of infrastructure spending. OpenAI has projected annual revenue of 200 billion USD by 2030, yet in the same period it expects cumulative server costs alone to reach 450 billion USD. Given the overwhelming share of capital required for server expansion, GPU acquisition, and data center construction, much of this will translate directly into cash outflow. This suggests that the operating cost structure needed to remain competitive in model development may continue to erode revenue gains, deepening losses rather than reducing them.

With both revenue and losses expanding simultaneously, OpenAI’s push to aggressively target the enterprise AI market is viewed as essential rather than optional. Dresser, who oversaw the Salesforce–Slack integration, has experience building stable B2B revenue pipelines. What OpenAI seeks from him is the ability to convert the uncertainty of “model competition” into predictable and recurring revenue. Within the industry, many interpret this as evidence that with model performance reaching a mature threshold, the competitive focus is shifting toward commercial execution.

Persistent debate over high valuation and lack of sustainable revenue models

Analysts increasingly view OpenAI’s deficit cycle as having reached a stage where short-term adjustments cannot resolve underlying issues. Infrastructure expansion and model development costs continue to rise at a pace far exceeding revenue growth, making external capital indispensable to day-to-day operations. Many recent investment rounds have been heavily front-loaded to secure computing resources and expand data centers, further extending the timeline until returns materialize. As financial pressure intersects with intensifying technological competition, many believe OpenAI faces even greater uncertainty in recouping its costs.

Concerns over inflated valuations and circular investment structures further cloud OpenAI’s long-term financial outlook. Debate around potential overvaluation intensified after reports last year of a proposed 30 billion USD investment from SoftBank. More recently, speculation surrounding supply-linked investment arrangements with Nvidia, AMD, and Broadcom has reignited these concerns. Critics argue that much of the capital supplied by investors flows immediately into semiconductor purchases and infrastructure buildout, leaving limited net cash flow and producing a more convoluted equity structure.

Delayed monetization of ChatGPT remains another serious challenge. Weekly active users already exceed 800 million, far surpassing market expectations. Yet paid subscribers remain below 5 percent, meaning the company has yet to establish meaningful cash flow. In an environment where price discipline is difficult, extended user engagement inevitably expands losses. One industry expert noted that “large-model operations require rapid early-stage monetization,” warning that without this, reducing losses would require substantial additional cost and time.

These factors lend weight to arguments that the “AI bubble” reflects not mere metaphor but a concrete financial dilemma stemming from OpenAI’s cost structure. Fears of falling behind in model competition, debate over the efficiency of heavy infrastructure spending, and long time-to-revenue are coming together to widen the gap between OpenAI’s valuation and its path to profitability. Major outlets such as the Financial Times have recently written that “if OpenAI maintains its current cost structure, sustaining operations without continued external capital may be virtually impossible,” further deepening market concerns.

IPO discussions accelerate amid mixed expectations

Amid rising scrutiny, OpenAI is pursuing an initial public offering as a means to reverse market sentiment. The company is reportedly targeting a valuation of 1 trillion USD, with an expected offering size of around 60 billion USD. To prepare for this, OpenAI entered into an agreement with Microsoft late October, finalizing a corporate restructuring plan to convert the company into a public-benefit corporation (PBC) under nonprofit oversight. Microsoft agreed to support the establishment of the PBC and the capital reorganization process, and would hold approximately 27 percent of its equity. This restructured governance framework is intended to bolster OpenAI’s prospects of securing substantial IPO capital.

Investment analysts believe OpenAI’s IPO could either deflate or crystallize the AI bubble narrative. Morgan Stanley described OpenAI’s deal-making structure as resembling a “plate of spaghetti,” arguing that such entanglement reveals the limits of circular transactions. While this intertwined network of supply chains and capital relationships may help OpenAI scale quickly in the short term, it also raises concerns about transparency and sustainability—two factors that public-market investors weigh most heavily.

Skepticism has also surfaced in Silicon Valley. According to Business Insider, OpenAI ranked second in an informal vote among roughly 300 founders and investors at the “Cerebral Valley AI Summit” in San Francisco last month on which AI company is most likely to collapse first. Aggressive investment combined with deepening losses has heightened anxiety, especially given projections that OpenAI’s cash burn rate will remain around 57 percent through 2027. Cash burn rate reflects how rapidly a company consumes the funds required for ongoing operations.

Yet in the same survey, OpenAI also ranked second on the list of “most desirable private companies to invest in,” underscoring the coexistence of optimism and caution. This polarity reflects the degree to which OpenAI’s IPO will shape broader market psychology. A successful listing would lend credibility to the company’s long-term growth roadmap and reinforce the narrative that AI is an industry capable of generating real profits. But if demand during the offering proves weak or valuations fail to meet expectations, mounting bubble concerns could gain further traction. As a result, OpenAI’s IPO is viewed as a critical turning point that could either ease or amplify the debate over an AI market bubble.