A fight over AI coding tools is making headlines. Anthropic, the company behind Claude Code, sent a legal notice to a developer who tried to share its hidden code. Meanwhile, OpenAI’s Codex CLI is winning praise for allowing everyone to use and improve its code freely.
Claude Code vs. Codex CLI: Two Ways to Build AI Tools
Both Claude Code and Codex CLI are AI tools that help developers write code faster. But they work very differently.
- Codex CLI (OpenAI): Anyone can see, change, and share its code. Developers even added features that allowed it to work with rivals like Anthropic.
- Claude Code (Anthropic): The code is locked and hard to read. When a developer shared it online, Anthropic used copyright laws to remove it.
OpenAI’s open approach helped it gain trust. Anthropic’s strict rules upset developers who want to learn from or improve the tool.

Why Developers Are Angry at Anthropic
Developers love tools they can tweak and study. When Anthropic forced GitHub to delete the shared Claude Code, many called it unfair.
What Happened on Social Media
On sites like X and Reddit, developers said OpenAI’s open style is better for innovation. One wrote, “If I cannot see the code, I cannot trust it.” Others praised Codex CLI for letting them add new features quickly.
Anthropic says Claude Code is still in testing and might change. But right now, its rules stop developers from exploring how it works.
What This Means for the Future of AI Tools
OpenAI’s CEO, Sam Altman, recently said the company was “wrong” to ignore open-source ideas before. Now, Codex CLI’s success shows how sharing code can win support.
Anthropic’s closed strategy might work for companies that want control. But individual developers and small teams prefer tools they can customize. For example, a developer in Brazil used Codex CLI to create a Portuguese coding helper, something Claude Code’s rules might block.

The Big Lesson for AI Companies
The conflict demonstrates expanding differences between companies. OpenAI joins other companies that endorse code sharing as a method to establish trust between developers and organizations. According to Anthropic, the release of coding tools as open-source might result in wrong or harmful utilization by users.
More people are currently adopting OpenAI’s current approach. Developer philosophy suggests that open-source tools create a team-oriented environment. Closed tools feel like rules.” Success will most likely go to the organization that pays attention to developers’ wishes, including open communication and adaptability, along with creative autonomy.