His Technologies AI BOTs & Educational Microworlds
His innovation began long ago. Biz has always been a Visionary. Trained by his father, the young Biz was a prodigy in Math, mastering Calculus by age 14. After graduating from MIT in 1985 where he was heavily influenced by the late Dr. Seymour Papert and Undergraduate Research Opportunities at the MIT Media Lab, Biz continued his relentless pursuit of Human AI solutions. At CUNY Queens College where Biz got his Master of Arts in Computer Science, he studied and published white papers about Autonomous Agents with Dr. Bon Sy.
In His Vision for STUFF, a Papertian MicroWorld of Chemistry, he saw the value of using simulations and graphics to show students how molecules and ions would react to each other or to heat for example. He knew that chemicals are dangerous for children and on top of that, many inner city kids wouldn't even have access to a bunsen burner. He knew and took on faith that an animated Digital World would evolve. Hence, some of these ideas he has conceived in the eighties won't even be fully realized until the Metaverse happens in the next few years (nearly 50 years later).
His vision for AI BOTs was apparent when in the nineties he published the Virtual Knowledge Architecture for Robot Planning. In this paper, Biz identified that you don't have to know something if you know where to get the knowledge. When he published the work on Discretion, the Smart Interactive Television solution in the nineties, Biz envisioned the future of Digital Streaming and assumed that everything Media would become Digital. He knew that TV shows, Movies, Music, Books, products of Information Systems and Web sites could eventually be delivered intelligently with the understanding of people and/or companies and their profiles. Based on the assumption, it was clear to him then as it is even more now, that large systems that serve Media to people will become Digital Document Management Systems.
Like Papert was, Biz is extremely metacognitive and uses introspection to model his AI BOTs so they may think like humans. In this sense, he is also an epistemologist but he goes a step further to cite that in addition he must also be a cultural anthropologist. He teaches that it is not enough to be able to know how you think (or how Americans think) about a problem and its solution, as much as whether the same problem exists in another culture and how people think about it there. He believes there is a such thing as "mental culture" - a way of thinking and problem solving which is influenced by physical immersion in a particular culture. He also teaches that "Artificial Intelligence without Epistemology is like Medicine without Biology."
His design AI BOTs was inspired by how ants collaborate to produce useful work. Today, these same ideas are being proposed in Cloud Native Artificial Intelligence solutions that use "Microservices" and were playfully depicted in BIG HERO 6 as Hiro Himada's first Robot Design. He fully endorses the current work by FETCH.AI and their "agent-aware ecosystem" concept. Furthermore, how FETCH, like Biz, also talks about Financial AI BOTs that work to help people gain economic advantages. His Enterprise is only beginning now, but his success lies comfortably in the Future he has enVisioned for years.
June 1985 (J.C.R. Licklider, thesis advisor)
“STUFF” – An Interactive Microworld Experience that Helps Children to Learn Chemistry.This educational learning environment enables students to play with chemical elements. It allows them to set up Laboratory Experiments, create chemical compounds, watch and zoom in to observe molecular activity, and study how molecules behave under varying temperature conditions.
IEEE Computer Society Press, 1993 September 22, 1993
The authors present an architecture for a system which will learn to anticipate and avoid problems,including both problems in accomplishing the overall goal of the system as well as problems normally associated with cooperation. The authors envision a future of autonomous robots which will be asked to accomplish a common goal as a team. The coach agent is a case based planner whose purpose is to learn employing cases of very successful teams and to anticipate and avoid unsuccessful teams.
IEEE International Conference on Systems, Man, and Cybernetics, Citations: (2) 1997
An agent-aware digital library architecture (termed DiscretionTM) for mediation of interaction between people and media information servers is proposed.Discretion is a digital library. It is organized as a hierarchical collection of objects and makes use of object oriented techniques for the classification, localization, and indexing of its contents. Discretion provides interfaces for the maintainer of the digital library (the librarian) to manage library content, for end users to indicate their overall preferences, and to enter requests for library items (or queries). It employs agents, virtual attorneys working on behalf of the user, to interrogate the Librarian when new items are being brought into the library. In this paper, two ways in which agents can assist end users are presented. The goal is to address the issues about organizing, filtering and personalizing data available to end users so that the gap between what users want and what they end up getting can be bridged. The work suggests that interactive television can be a way to test the interface agent technology and to prove its use.
MyBOTS.Tv—the personal helper BOT system utilizes a customizable personal assistant in the form of an avatar to engage its owner in a conversational manner with natural language commands to coordinate activity of a team of autonomous helper BOTs.Among a variety of other tasks, BOTs, which are software applications that run automated tasks over the Internet, collaborate to create the owner's schedule, maintain his/her to-do list, obtain personally interesting information, provide personalized services and searches, perform web site transactions, use all types of web apps, complete tasks and/or synthesize useful products for their owners.
In some embodiments, the helper BOTs system leverages Digital Library (DL) Architectures to house elements that they intelligently serve. The underlying Digital Library Architecture features an induction process that elicits facts about incoming elements that can be judged by a panel of agents. Agents judge elements according to areas of discretion which are of interest to the consumer population. The judgments are then stored as part of each element's metadata. Helper BOTs then deliver the elements intelligently by probabilistically ranking, sorting, classifying, and/or presenting and serving them to users in accordance with value preferences indicated in their profiles. One such embodiment, (which has been previously published as a Virtual Knowledge Architecture for Interactive Television) ranks and serves TV programs to people based on the value preferences set in their profile(s).
In other embodiments, helper BOTs leverage user-directed search paradigms to refine and customize the search function according to the searcher's preferences, style and demographic information. In these embodiments, helper BOTs are directed by way of a search construction wizard to return search results in accordance with searcher profiles and demands such as language, context of content, age of document, and age appropriateness. In this regard, searcher preferences refer not only to the search itself, but to the style, presentation, and appropriateness of the search results which can not only be a list of links but delivered as synthesized living document products incorporating new media elements.
Some helper BOT embodiments will help people to perform financial transactions requiring highly sensitive and personal information such as credit card numbers, bank account information and the like. These embodiments require the use of BOT security. BOT security will utilize x.509 digital certificate technology assigned to the avatar by its human owner. The cryptographic key will be utilized to perform all related transactions and protect any information deemed sensitive on behalf of its owner in a secure manner through secure network tunnels.
Further embodiments enable an extensible set of commercially available BOTs to be obtained on the MyBOTs.Tv web site in the BOTS Store. This has begun with the 12 “EmBOTiments” presented here. These BOTs perform individual tasks autonomously but may also work in collaboration with each other. Each BOT processes inputs and can be parameterized. Hence, a BOT API enables the creation of stored programs that capture the teamwork interaction and makes their function re-usable. An extensible grammar makes it possible for the evolving baseline of useful stored programs to be called by name so that BOT teams can be dispatched by the user's avatar on demand.