Oracle bans AI-generated code from OpenJDK

Oracle wants AI writing its own software, but if you contribute to Java, keep the LLM out of the pull request.

Oracle has banned AI-generated code from OpenJDK contributions, citing safety, security, and intellectual property risks, according to a report from The Register. Developers can still use LLMs privately to debug or review code. However, submitting AI-generated material to repositories, pull requests, or official project channels is strictly prohibited.

The rule sharply contrasts with Oracle's internal operations. Co-founder Larry Ellison recently declared that AI models now write Oracle's code, while co-CEO Mike Sicilia credited AI tools with enabling smaller teams to deliver faster. Meanwhile, Oracle is spending $70 billion on data center expansion this year—a spree that prompted credit agency S&P to downgrade Oracle's rating to BBB-, one notch above junk status, citing uncertain returns.

Why it matters: OpenJDK forms the core of Java development, and Oracle cannot afford copyright uncertainty or unverified code in an enterprise foundation. Internal productivity hacks are fine for Oracle's bottom line, but public codebases get zero-tolerance legal scrutiny.

Know this: You can still use AI tools to review or debug your OpenJDK contributions locally. Just make sure the final code hitting the commit is human-written.

Expect other enterprise open-source stewards to copy this policy soon.