Claude, Codex rivalry fuels AI coding debate

Revenue is a business signal, not a capability scoreboard. Anthropic’s coding products have strong momentum, but the leap from “Claude is monetizing well” to “OpenAI is finished” is not supported by the evidence.

• –Semafor notes Anthropic and OpenAI account for cloud-partner revenue differently, so leaked numbers are hard to compare cleanly.
• –Axios reports Anthropic’s revenue and usage grew 80x year over year in early 2026, which explains why Claude Code keeps getting framed as the hot hand.
• –OpenAI still has Codex, broad ChatGPT distribution, and a much larger consumer footprint, so the competitive picture is not one-way.
• –For developers, task fit matters more than brand tribalism: long-running codebase work, workflow integration, and pricing usually decide the winner.