On October 29, BotID Deep Analysis detected a 500% traffic spike that proved to be a sophisticated bot network with 40-45 profiles mimicking legitimate users. The system, powered by Kasada's ML backend, automatically adapted within 10 minutes to identify the coordinated proxy-cycling pattern and reclassified the traffic as malicious without manual intervention. This case study demonstrates real-time adaptive detection of zero-day bot networks combining real browser automation and legitimate-looking telemetry.
Safety
BotID Deep Analysis catches a sophisticated bot network in real-time
Kasada's BotID autonomously detected a sophisticated 45-profile bot network mimicking a 500% traffic spike in 10 minutes, automatically adapting to proxy-cycling evasion tactics without manual intervention.
Friday, April 17, 2026 12:00 PM UTC2 MIN READSOURCE: Vercel BlogBY sys://pipeline
Tags
safety
/// RELATED
InfrastructureApr 17
How Nous Research used BotID to block automated abuse at scale
Nous Research integrated Vercel's BotID bot detection to block thousands of automated fake accounts that exploited their free Hermes LLM tier during a promotional week, preventing similar abuse when reopening free offerings.
SafetyApr 17
BotID uncovers hidden SEO poisoning
Legacy SEO poisoning attacks resurface as Google crawlers re-index years-old compromised URLs, caught by Vercel's BotID bot detection—showing how historical security breaches can silently generate fraudulent traffic until rediscovered.