IBUKI: Building AI Agents
 


IBUKI: Building AI Agents

IBUKI Products

IBUKI products have a common theme: they all relvolve around sotware for enabling AI in applications.

Our products uses ideas that were well discussed 50 year ago when the most powerful computers were less powerful than today's smartphones. What generative AI using LLMs has taught us is that massively scaling old algorithms (e.g., neural nets and tenor algebra) using the latest hardware make a qualitive difference in functionality.

It is not that IBUKI is buried in the past and does not recognize both the intellectual and commercial power of generative AI. Instead, IBUKI is focused on preparing its customers for what comes next.

NEWFOL

NEWFOL is a development framework for building AI agents. It consists of two parts: tools for developing the interior of an agent—that includes making and modifying an agent's state (IBML types); and, tools for writing programs that sit on an agent's boundry and interact with its exterior—programs that communicate with other agents or use sensors and affectors to understand its world.    find out more >>

NEWFOL Cognitive Models

An AI application without any knowledge wouldn't be very intereting. NEWFOL models an agents cognitive state with data structures called [context]s. Each [context] is a self-contained database that contains information about some subject. The NEWFOL Framework comes with some built-in [context]s but the IBUKI Database contains over ten million of these 'mental' models that can be licenced for use in building AI applications. [context]s contain both 'common ground' knowledge and specialized subject knowledge.    find out more >>

Other Software

Fred is not really a product but rather a live example of an AI agent who is able to have conversations about the normal things of life (common ground) in English. He used NEWFOL [context]s to store his knowledge and uses rule-base semantic NLP software to both understand and peak English.    find out more >>