AI coding wars
A recent column says OpenAI, Google and Anthropic are no longer just releasing models — they’re competing directly over coding workflows and developer tooling. The piece argues that the battle has shifted from consumer chatbots to tools that shape how engineers write, test and ship code. (theverge.com)
OpenAI, Google, and Anthropic have shifted their fight from consumer chatbots to directly competing over coding workflows and developer tooling. (theverge.com) OpenAI launched GPT‑5.3‑Codex and a Codex desktop app in February 2026, pitching an agentic coding model that it says scores highly on SWE‑Bench Pro and Terminal‑Bench 2.0. (openai.com) Anthropic rolled out Claude Opus 4.6 on February 5, 2026, added a 1 million‑token context window and introduced multi‑agent features and a "Code Review" tool inside Claude Code. (anthropic.com) Google has packaged Gemini into Gemini Code Assist and Vertex AI integrations, offering in‑IDE assistants, enterprise customization, and Cloud Shell preconfigured environments. (codeassist.google) This race traces back to GitHub Copilot’s June 29, 2021 technical preview, which first put AI suggestions inside the IDE and started developers shifting to assistant‑driven workflows. (github.blog) Adoption signals include GPT‑5.3‑Codex rolling into GitHub Copilot in February 2026 and engineers at multiple firms saying AI now generates the bulk of routine code. (github.blog) Companies are responding with product and pricing moves: OpenAI added a $100/month ChatGPT Pro tier aimed at heavier Codex users in April 2026, while Anthropic and Microsoft push enterprise Foundry and controls for Claude. (techcrunch.com) The shift has produced friction: security and source‑code exposures tied to Anthropic’s Claude Code leaked recently, prompting takedown requests and developer scrutiny. (arstechnica.com) The Verge column framing this as a move from chatbots to developer tooling says the winner will be whoever embeds AI into the everyday tools engineers already use. (theverge.com)
