Who’s in control: Human or Machine?
summary
Detects bots by specific signals not buzzwords: device and IP intelligence plus behavioral biometrics (mouse trajectories, velocity, acceleration, keystroke timing, touch points, motion sensors), sampled smartly (typical windows ~1–2 second sampling) and fingerprint replay-detected; defenders deploy proof-of-work via JavaScript and ML to correlate context and raise cost. Attackers evolve from dozens to hundreds to 10,000+ node botnets, spoof fingerprints, replay or inject noise, or use headless browsers, so the integrated system works to raise attacker cost but demands heavy data engineering and ongoing tuning.
AI-written summary · about
chapters (16)
- 0:00Intro
- 2:22Types of attack
- 7:46What drives attack sophistication
- 9:34Evolution of attack and detection methods
- 16:43Behavioral detection: User Behavior Analysis
- 17:40Behavioral detection: Behavioral biometric detection
- 18:35Behavioral Detection Use Cases
- 19:37User Behavioral Analysis Examines the context of the interactions
- 22:30Behavioral biometrics available depends on the client
- 27:23Behavioral biometrics - Mouse Motion Timing
- 28:43Behavioral biometrics - Mouse Velocity
- 29:47Behavioral biometrics - Mouse Acceleration
- 30:54Behavioral biometrics - Touch interactions Machine
- 33:13Behavioral biometrics - Key Press Metrics
- 35:15Behavioral biometrics - Spoofing • Many different ways to falsify inputs
- 38:23Why use behavioral detection? Pros
related talks
23mBot or human? Detecting malicious bots with machine...
31mStrategies for Identifying and Mitigating Bot Threats
29mStandard Classification: Good Bot, Bad Bot: Characterizing Automated Brows...
01h 00mRock ‘Em Sock ‘Em Robots: Bot Swatting Like the Pros
12mSpeed, scale & stealth: Securing against ATO events
49mAre you human or a damn bot?
38mRise Of The Machines - How Automated Processes Overtook the Web
40mRock 'Em Sock 'Em Robots: Bot Swatting Like the Pros