Honey Terminal
Disruptive Deceptive Cyber-Defense
Introduction
HoneyTerminal is the deceptive agent living inside your network enticing bad actors to access a "machine" that maybe too good to be true. HoneyTerminal leverages Generative AI and controls the "infamous" hallucinations to masquerade and respond as if it was as a critical network asset but in reality the bad actors are interacting with a never ending source of "honey" and misinformation. HoneyTerminal can misdirect, misinform, and deceive bad actors dynamically, allowing the breach to be contained before actual damage can occur.
Problem
Detecting and responding to cyber attacks is extremely challenging and often a breach is realized after significant damage is already done. "Honeypots", an exposed machine with fake information held within, can be deployed in a network to attract, monitor, and contain the bad actor before they progress to the “real” network. The challenge comes in maintaining, configuring, and supplying “honey” to the bad actor. Where typical honeypots are static and do not change, Generative AI supported HoneyTerminal allows honey data, system behaviors, and characteristics to be generated on the fly. HoneyTerminal can trap the bad actor within it indefinitately.
Opportunity
Our initial prototype acts as if it was a standard Ubuntu shell terminal. Users, home directories, and file systems are configurable and the HoneyTerminal can be traversed through as if it was a real system. A description of the type of business and characteristics of the “honey” the company may have and the AI model will generate files, data, and users as if it is a real machine. Future efforts will be into formalizing the definitions and descriptions of the HoneyTerminal inputs to enable customers to quicky define and spin up HoneyTerminals, making it more challenging for bad actors to identify the real from the fake.