Crypto platform Gemini has reduced its workforce by about 30% since the start of 2026, saying the move reflects a deep integration of artificial intelligence into its operating model. In a shareholder letter, management argued that the era of the “10x engineer” is over, claiming that employees equipped with AI agents can now deliver a “100x” impact across both technical and non-technical functions.
AI adoption is accelerating fast
According to the company, the transition has moved quickly. During Gemini’s IPO roadshow in late 2025, only 8% of production code was generated or assisted by AI. By December, that trend had reached an inflection point. Gemini now says more than 40% of production code changes are AI-generated or AI-assisted, and it expects that share to approach 100% in the near future. Management framed the shift bluntly, saying that not using AI at Gemini will soon be like showing up to work with a typewriter instead of a laptop.
The company argues that a leaner structure supported by high-leverage AI tools is not simply smaller, but faster and more agile. That logic underpins why Gemini is presenting layoffs and productivity gains as part of the same strategic transition.
Revenue growth clashes with worsening losses
Still, Gemini’s AI pivot comes against a financially mixed backdrop. The firm said fourth-quarter 2025 revenue reached $56.4 million, its highest quarterly figure in three years and up 13% from the previous quarter, even though spot trading volumes fell 30%. Full-year net revenue came in at $174.1 million.
Those gains were overshadowed by steep costs. Operating expenses in Q4 climbed to $171.7 million, more than triple net revenue for the period. For the full year, “other expenses” saw a $258 million year-over-year swing, moving from a $14.9 million gain in 2024 to a $243.1 million loss in 2025. As a result, Gemini’s total FY2025 loss widened to nearly $585 million, compared with $157 million in 2024.
The key question for Gemini is whether AI-led productivity gains can eventually outweigh rising costs and large-scale losses. For now, the company can point to improving revenue, but its financial disclosures suggest it is still far from proving that an AI-first strategy can fully repair the bottom line.