In a recent LinkedIn post, Linas Beliūnas analyzes a bold new strategy reportedly being employed by OpenAI, drawing parallels to financial maneuvers and the early days of disruptive tech companies. Beliūnas frames OpenAI’s current approach not as a simple sale of artificial intelligence, but as a sophisticated play to finance its own widespread adoption and secure long-term enterprise dominance.
Beliūnas highlights the significant financial underpinnings of this strategy, noting the reported substantial annual losses faced by the AI giant. He points out that despite these losses, OpenAI is offering investors what appears to be a highly attractive, “guaranteed” 17.5% return through joint venture vehicles pitched to private equity firms.
“OpenAI is losing $14 billion a year, but it’s now offering investors a ‘guaranteed’ 17.5% return. Read that again 😳”
According to Beliūnas, these aren’t pitches for core equity but rather for ‘side vehicles’ designed to roll out AI across a vast number of portfolio companies. He characterizes this as essentially “paid distribution.”
The Shift from Intelligence to Integration
A central theme in Beliūnas’s analysis is the evolving bottleneck in the AI landscape. He argues that the primary challenge is no longer the intelligence of the AI models themselves, but rather their effective integration into existing business processes. This integration, as Beliūnas points out, is inherently difficult.
“Integration is slow, messy, and expensive.”
Beliūnas elaborates that OpenAI is currently operating at an impressive enterprise revenue run rate, estimated to be over $10 billion. However, this is still overshadowed by massive losses driven by compute and infrastructure costs. He also notes the increasing competition, with Anthropic reportedly gaining traction in the enterprise market.
Financing Rollout and Locking in Users
Beliūnas posits that OpenAI CEO Sam Altman has fundamentally altered the business model. Instead of merely selling AI technology, the company is now actively financing its deployment. This involves turning potential buyers into partners, and subsequently, into deeply entrenched, locked-in users.
“OpenAI is now at ~$10B+ enterprise revenue run rate → But losses still massive (compute + infra) → Anthropic gaining ground in enterprise”
He draws a parallel to the early strategy of companies like Uber, which Beliūnas describes as a playbook of “lose money upfront → own the network later.” In this context, the AI race has shifted from a focus on model superiority to a critical battle for distribution control.
The Power of Embedded AI
The core of OpenAI’s long-term bet, as interpreted by Beliūnas, lies in the difficulty of removing AI once it becomes deeply integrated into a company’s operations. He suggests the real value is not just in creating smarter AI, but in making it indispensable.
“The AI race has shifted from ‘who has the best model’ → to ‘who owns distribution.’ … Because once AI is deeply embedded, switching becomes nearly impossible.”
Beliūnas concludes that this strategy aims to subsidize adoption, thereby securing enterprise usage and creating a powerful network effect that competitors will find extremely difficult to overcome. The focus is on creating dependencies that ensure future revenue and market share, regardless of the immediate profitability of the AI models themselves.
📝 About This Content
This article is based on insights shared by Linas Beliūnas on LinkedIn.
📅 Originally posted on March 23, 2026 | View original post on LinkedIn →