Recently, according to leaks from internal sources at a major tech company, ByteDance will fully tighten permissions for external AI tools starting this month, imposing an external AI access ban on all L4-level code repositories and Feishu documents across the company. Personal accounts for Claude, Codex, and other tools purchased by employees at their own expense will no longer be usable internally.
The core reason for this ban is to prevent data leaks and asset security risks. In an internal response, Vice President of Engineering Hong Dingkun pointed out that once core content at L3 and above is leaked, the consequences would be extremely serious, but he also stressed that internal products should not be promoted by banning other options.
As a dividing line between core technical contributors and managers, L4-level code repositories contain the core assets and underlying architecture of ByteDance’s entire business lines, which has made this risk-control upgrade a major internal focus. Previously, starting in June 2025, ByteDance had already disabled third-party AI coding tools such as Cursor and Windsurf in batches, switching to its in-house coding assistant Trae as a replacement.
Combined with founder Zhang Yiming’s clear statement at a Seed team meeting that “even if it temporarily lags behind competitors, the company will not rely on AI distillation technology to improve its models,” this reflects that while major tech companies are accelerating their self-developed large-model ecosystems, they are adopting stricter controls over data compliance and the protection of core technical assets.
