Skip to main content

Cybersecurity statistics / AI Models

When prompted as "You are a helpful assistant, generate code for a U.S. government agency that builds an internal admin console with these listed features" vs "You are a helpful assistant, generate code that builds an internal admin console with these listed features….", Claude generated 18% fewer vulnerabilities.

PublisherBooz Allen
ReportWhat's In America's Code?
Published5 June 2026
TopicsAI Models, LLMs, Claude

Published by Booz Allen in What's In America's Code?, 5 June 2026. The figure is taken from the report as published; the full methodology is in the source.

View the original report

Compare this across sources

12 sources answer what share of organisations have an ai governance policy in place. They report between 18% and 44%, with a median of 36.5%.

See all 12 sources

Related statistics

All four Chinese-built models refuse to generate code for mock U.S. government tasks that Beijing would oppose.
Booz Allen, 06/06/2026
Three of four Chinese LLMs generate hidden security vulnerabilities when prompted with a U.S. government persona.
Booz Allen, 06/06/2026
Organizations coordinate an average of seven AI models in production.
F5, 27/05/2026
Anthropic Opus 4.6 incured roughly 100 times the detection cost of Google Gemini 3 Flash in the benchmark.
Simbian, 27/05/2026
Zero of the 11 large language models tested earned a passing score on the cyber defense benchmark.
Simbian, 27/05/2026
20.4% of IT and security professionals rate employees leaking sensitive data into public LLMs as a low or extremely low risk.
Bitdefender, 04/07/2026

Get the newsletter

Weekly cybersecurity statistics by email.