How Anthropic secures its AI-native software development lifecycle

Jason Clinton, Anthropic’s deputy CISO, on what happens to security review when code volume jumps and the old checkpoints turn into bottlenecks. His starting numbers: Claude authors about 80% of the code merged into Anthropic’s codebase, more than half of it merged by their internal version of Claude Tag with engineers setting intent and owning final approval, and engineers ship 8x as much code per quarter as they did from 2021 to 2025.

The pipeline he describes runs across all five stages rather than sitting at one gate. Automated project security reviews at planning; secure coding practice encoded in CLAUDE.md files and remote VMs with egress allowlisting at the coding stage; several specialised review agents on each PR in CI, with human approval still required for regulated and critical code; continuous AI-driven DAST scans against staging; and a single-purpose, restricted-permission agent doing alert triage in production. The consistent theme is hard access boundaries and identity containment rather than trusting instructions in a prompt.

Two things worth keeping. The share of PRs getting substantive review comments went from 16% to 54%, which grew as they made agents prove their findings were real rather than just flagging them. And his framing of where the job goes:

The security engineer’s job evolves from monitoring bugs to monitoring loops.

Anthropic writing about the benefits of Anthropic’s own tools, so read the numbers accordingly.