Within the battle between two “agentic” coding instruments — Anthropic’s Claude Code and OpenAI’s Codex CLI — the latter seems to be fostering extra developer goodwill than the previous. That’s not less than partly as a result of Anthropic has issued takedown notices to a developer attempting to reverse-engineer Claude Code, which is below a extra restrictive utilization license than Codex CLI.
Claude Code and Codex CLI are dueling instruments that accomplish a lot of the identical factor: permit builders to faucet into the facility of AI fashions working within the cloud to finish numerous coding duties. Anthropic and OpenAI launched them inside months of one another — every firm racing to seize worthwhile developer mindshare.
The supply code for Codex CLI is obtainable below an Apache 2.0 license that enables for distribution and industrial use. That’s in distinction to Claude Code, which is tied to Anthropic’s industrial license. That limits how it may be modified with out specific permission from the corporate.
Anthropic additionally “obfuscated” the supply code for Claude Code. In different phrases, Claude Code’s supply code isn’t available. When a developer de-obfuscated it and launched the supply code on GitHub, Anthropic filed a DMCA criticism — a copyright notification requesting the code’s removing.
Builders on social media weren’t happy by the transfer, which they stated in contrast unfavorably with OpenAI’s rollout of Codex CLI. Within the week or so since Codex CLI’s launch, OpenAI has merged dozens of developer strategies into the device’s codebase, together with one which lets Codex CLI faucet AI fashions from rival suppliers — together with Anthropic.
Anthropic didn’t reply to a request for remark. To be truthful to the lab, Claude Code continues to be in beta (and a bit buggy); it’s potential Anthropic will launch the supply code below a permissive license sooner or later. Firms have many causes for obfuscating code, safety concerns being one in all them.
It’s a considerably shocking PR win for OpenAI, which in latest months has shied away from open supply releases in favor of proprietary, locked-down merchandise. It might be emblematic of a broader shift within the lab’s method; OpenAI CEO Sam Altman earlier this 12 months stated he believed that the corporate has been on the “unsuitable facet of historical past” relating to open supply.
