HUMAN is thrilled to announce a generational milestone in our detection, mitigation, and reporting capabilities. These new AI-enabled capabilities enhance HUMAN’s core decision engine, coming on the heels of our Leader recognition in The Forrester Wave™: Bot Management Software, Q3 2024 and our position as #1 in the G2 summer grid.

This release marks a significant enrichment to the Human Defense Platform. A set of targeted machine learning models layer across our core decision engine to analyze fraudulent traffic and optimize mitigation rules after a bot-or-not decision is made.

How It Works

Interconnected machine learning models automatically analyze the malicious traffic that is detected on your specific application(s), looking at the large-scale patterns in the traffic in order to identify and track individual attacker profiles. Other models monitor these profiles and automatically execute new mitigation workflows based on the unique characteristics of the fraudulent traffic and account activity, even as the attacker attempts to pivot. 

Here’s the process:

1. HUMAN’s industry-leading decision engine analyzes traffic to detect and mitigate sophisticated bots. 
2. Once a bot-or-not decision is made, new machine learning models ingest various characteristics of the malicious traffic in aggregate to profile threats, provide insights on attack characteristics and attacker actions, and identify commonalities that connect seemingly disparate attacks.
3. Using these patterns and profiles, other models automatically adjust response flows to mitigate threats as they evolve.

Together, these machine learning models form a larger system of artificial intelligence (AI) that enables enhanced detection, mitigation, and reporting. 

New Capabilities

With HUMAN’s enhanced decision engine, the following AI-derived capabilities will be available in Account Protection and Application Protection.

• Adaptive Detection: Tracks and blocks attackers as they adapt and change tactics over time, ensuring a continuous line of sight and continuous protection even after the initial decision is made
• Attack Profiling: Segments malicious traffic into distinct attack profiles and provides visibility into attack characteristics and attacker actions, allowing analysis of attacks beyond just traffic spikes 

• New Dashboard & Investigation Tools: Delivers intelligence that allows customers to jump straight into their analysis, saving time and improving understanding of what exactly happened during an attack 

• Network Attack Event Detection: Helps customers understand the scale and complexity of an attack to better identify and neutralize large-scale abuse of fake and compromised accounts

• Profile Deviation ML Models 2.0: Enhances detection precision of malicious post-login activity, so our customers can catch even the most stealthy cybercriminals who attempt to mimic the real account holder’s behavior

With these AI-derived capabilities, HUMAN goes far beyond bot detection and mitigation to also identify and surface key details on hidden attack patterns and profiles for the analyst. You can see that a specific bot was active on your site, how it was mitigated and also understand exactly what the attacker was attempting to do and how this attack is related to other attacks. 

This enables our customers to uncover hidden threats, hit the ground running on incident analysis, and respond faster to evolving threats. 

Building on Unparalleled Data Visibility

HUMAN's core decision engine has always been powered by the most advanced set of statistical and machine learning models in the industry. (Don’t believe me? Check out the Forrester Wave™: Bot Management Software, Q3 2024, where HUMAN was named a Leader and received the highest score possible in Detection Models before this announcement.)

HUMAN’s decision engine brings together signals from across the entire digital customer journey. We see more than 20 trillion digital interactions each week across 3 billion unique devices monthly. These interactions span advertising, application, and account surfaces. The Human Defense Platform analyzes the signals we gather through 400+ machine learning algorithms to detect fraud-or-not with unparalleled scale, speed, and decision precision.

Our enhanced system of machine learning models interplays with our core algorithms to reveal previously unseen insights into individual attack characteristics, attacker actions, and large-scale attack patterns. As a result, customers are able to conduct in-depth analysis of attacks and automatically optimize mitigation workflows in real time when a bot-or-not decision is made.

Continued Investment in AI Techniques

HUMAN’s AI system continuously ingests new attack data and learns over time. This strengthens our decision engine, laying the foundation for innovations that enable our customers to optimize a targeted defense strategy. Through the smart application of AI, HUMAN provides enhanced detection, mitigation, and reporting to deliver industry-leading protection against the most sophisticated cyberthreats. HUMAN is actively investigating and testing additional AI systems for enhancing detection capabilities.