OpenAI made its most aggressive competitive move yet against Anthropic, launching GPT-5.6 models and a new enterprise product called ChatGPT Work. The dual announcement signals OpenAI shift from pure capability competition to enterprise value delivery, betting that superior price-to-performance will win over corporate customers currently favoring Anthropic.

The market has flipped. Ramp AI Index for May 2026 showed Anthropic at 34.4 percent of business adoption against OpenAI 32.3 percent, the first time Anthropic led on that survey. Enterprise spend tilts toward Anthropic at 32-40 percent depending on measurement, down from OpenAI dominance at 50 percent in 2023. OpenAI product leadership is accelerating to stop the bleed.

The Pricing Angle

OpenAI response acknowledges what enterprise buyers actually need once pilots graduate into production systems. Anthropic leaned into reliability and instruction-faithful work. OpenAI counters with price cuts and productivity features. GPT-5.6 Luna beats Claude Opus 4.8 on some benchmarks while costing less, threatening Claude core value proposition.

Anthropic launched Claude Fable 5 at half the price of Mythos Preview just days before OpenAI price cuts, suggesting both companies anticipated competitive pricing pressure. The AI market is entering true competition after years of supply scarcity.

Enterprise Implications

Multi-model orchestration becomes real. Enterprises now compare not just capability but ROI on different model families. ChatGPT Work positions OpenAI as an alternative to Anthropic, while Anthropic continues building enterprise trust through reliability focus. The war is won not by benchmark dominance but by production performance and cost efficiency.

GPT-5.6 public availability marks first major AI pricing competition at frontier model tier.