OpenAI projects that ChatGPT Go, its lower-cost consumer tier priced at $8 a month in the United States, will reach 122 million subscribers by the end of 2026. The forecast, reported by The Information, signals a fundamental shift in how the company plans to generate revenue.

ChatGPT Go launched as a cheaper alternative to the $20-per-month ChatGPT Plus plan. In markets such as India, the tier costs roughly $5 a month. At the start of the year, OpenAI's internal projections estimated 112 million Go subscribers by December. That figure has since been revised upward to 122 million.

The strategy carries a deliberate trade-off. OpenAI expects tens of millions of existing Plus subscribers to downgrade to the cheaper plan. The company is betting that a much larger pool of lower-paying users, combined with advertising revenue, will more than compensate for the per-user revenue decline.

For three years, the $20 Plus subscription has been OpenAI's primary consumer revenue engine. Moving to a model that blends cheaper subscriptions with ad support marks a notable departure. It also brings OpenAI closer to the economics of consumer internet platforms, where scale and attention matter more than premium pricing.

The ad-supported component is key. OpenAI appears to be wagering that showing advertisements to a far larger user base will generate more total revenue than the existing higher-priced subscription alone. The company has not disclosed specific ad-revenue targets.

What this signals

The pivot reflects a broader recognition that $20 a month is a barrier for most consumers worldwide. Pricing ChatGPT Go at $5 in lower-income markets suggests OpenAI is pursuing global scale aggressively. A 36-fold increase in Go subscribers within a single year, if achieved, would represent one of the fastest consumer subscription ramp-ups in recent tech history.

It also raises questions about margins. Inference costs remain substantial for frontier models, and serving 122 million subscribers on a cheaper plan will pressure unit economics. Whether advertising fills that gap depends on engagement depth and advertiser appetite for AI-chat placements, neither of which is proven at scale.

Competitors such as Google with Gemini and Anthropic with Claude offer free tiers but have not yet moved aggressively into ad-supported AI chat. OpenAI's bet may force their hand.

The revised 122 million target suggests early adoption data has exceeded internal expectations. How retention holds through the year will determine whether the model works.