OpenAI plans to expand from roughly 4,500 employees to 8,000 by the end of 2026, according to the Financial Times, citing two people familiar with the matter. The hiring will span product development, engineering, research, and sales.

One notable addition: a new category of “technical ambassador” hires tasked with helping businesses integrate OpenAI’s tools. The role signals that OpenAI’s enterprise sales motion is shifting from self-serve API access toward hands-on deployment support, the kind of high-touch engagement that Anthropic has been winning with Claude.

Why Now

The timing is telling. According to corporate expense platform Ramp’s AI Index for March 2026, businesses are now 70% more likely to choose Anthropic over OpenAI when purchasing AI services for the first time. That number would have been unthinkable a year ago, when OpenAI held a commanding lead in enterprise adoption.

OpenAI’s response has been to buy market share on two fronts. On the government side, it secured a Department of Defense contract in February after Anthropic’s public falling-out with the Pentagon. On the private sector side, it’s in advanced talks with private equity firms including Brookfield Asset Management to deploy AI tools across their portfolio companies, Reuters reported.

3,500 Hires for a Company That Isn’t Profitable

OpenAI remains unprofitable. The company has not disclosed when it expects to reach break-even, and CEO Sam Altman has consistently framed the current period as one of necessary investment. Adding 3,500 employees at an average fully-loaded cost of $300,000 per head (a conservative estimate for Bay Area AI talent) would add over $1 billion in annual payroll alone.

The hiring push also lands in the middle of OpenAI’s complicated corporate restructuring. The company is still navigating its conversion from a nonprofit to a for-profit entity, a process that has drawn scrutiny from California’s attorney general and from Microsoft, which is reportedly weighing legal action over a separate Amazon cloud deal.

What 8,000 People Build

For context: Anthropic has roughly 1,500 employees, according to TechCrunch. Google DeepMind has around 3,000, per Wired’s reporting. Meta’s AI research division, FAIR, operates with fewer than 1,000, according to The Verge.

If OpenAI hits 8,000, it would be by far the largest dedicated AI lab in the world by headcount. The question is whether that size translates into faster product shipping or whether it creates the kind of organizational drag that smaller, more focused competitors avoid. Anthropic’s Claude has been gaining enterprise share with a fraction of the workforce, and much of that growth has come from a reliability and safety reputation that doesn’t require more engineers to maintain.

The “technical ambassador” role is perhaps the most revealing detail. OpenAI is acknowledging that API access alone doesn’t win enterprise deals. Companies want someone to show up, understand their stack, and help them deploy. That’s a services business, and services businesses scale with headcount, not with models.

Sources: Financial Times, Engadget, CNBC, Reuters