EFTA01084196.pdf
dataset_9 pdf 150.2 KB • Feb 3, 2026 • 3 pages
Iledee Funder Jeffrey Epstein Launches the First Emotional Robots
Thanks to a virtuoso band of scientists in Hong Kong and an unusual hedge hinder with a zealous science
background, called Jeffrey Epstein, the virtual world is moving away from algorithmic robots to a twilight
realm of emotional thinkers.
The Artificial Intelligence (AI) group in Hong Kong is called Open Cog. Open Cog develops
programming language for the AI community to use and share around the world. Much of Open Cog's
goal is to unify the fragmented field and language of Al, to accelerate research. However, Open Cog has
also been developing its own Al research: notably, shifting computer intelligence away from linear
algorithmic responses, towards goal driven, non-deterministic, emotional entities.
"Creating emotional Al models allows scientists to test their hypothesis about the mind," Jeffrey Epstein
remarked, the financial guru behind this effort. "The remaining gaps between these models and our
experience of the mind, is an invaluable guide to follow. It's somewhat like building a car, with no
instructions, but a visual of what a car can do."
Over the last ten years, Jeffrey Epstein has become one of the largest backers of cutting edge science
around the world. According to New York Magazine and National Review, he has donated up to $200
million a year to eminent scientists, including: Stephen Hawking, Marvin Minsky, Eric Lander, George
Church, and Nobel laureate physicists Gerard 't Hooft, David Gross, and Frank Wilczek. In 2003, Epstein
founded the Program for Evolutionary Dynamics at Harvard University, with a $30 million dollar gift to
the university. The Program studies the mathematical evolution of micro-biology and has made key
discoveries into the treatment of cancer, HIV and other infectious diseases.
For years, in the world of virtual gaming, the only intelligent player was the person playing the game,
responding to non-reacting obstacles. At most, upon virtual contact, obstacles could blow up, open up,
cave in, or morph into something else. Whatever the reaction, it was a simple linear or algorithmic
response (if A, then B, if A+D, then C).
Opponents soon became more complex with the development of such games as virtual chess, where the
computer responded to a vast network of algorithmic possibilities: up to 10123 game variations to be
exact. But even in those scenarios, the computer program does not have an ultimate goal: it does not aim
for check mate, but simply responds to a series of steps that lead to that direction.
EFTA01084196
Today's gaining characters from virtual soldiers to Tinkerbell, are also vastly more complex than their
dash line tennis forbearers. Like the chess interpreter, virtual soldiers can react to a wide variation of
landscape scenarios and in a myriad of ways.
Until now, emotional, needs-driven robots whether virtual or physical, were still in the realm of science
fiction.
Notes from here on...
Open Cog's virtual characters can.... There should be technical notes here on how their models work.
Also—there are virtual ones and actual physical, human sized robots. See PDF. Its very dense language.
• communicate via speech or text, empathic communication using biometric devices...
• The project supports the Hong Kong digital entertainment industry by supplying a toolkit for
developers to incorporate intelligent characters into their products.
• The prototype game world is built in Unity 3D, inspired by Minecraft. 3 characters: robot, girl
and ghost.
• Novel cognitive abilities,
• Using existing Open Cog software—building on that to create animated intelligent characters for
use in games, toys applications.
• Current technology allows the characters to have basic learning and reasoning, to communicate
in natural language about particular contexts.
• The Project supports the Hong Kong digital market by supplying a toolkit allowing developers to
incorporate intelligent characters into their products.
• The hypnosis that cognitive synergy among learning mechanisms associated with different types
of memory is valuable for embodied, human like general intelligence.
• Using Fishgram Algorithm: the capacity to recognize patterns in heterogeneous, rapidly
changing data. Language comprehension: semantic processing. Translating English into cognitive
content "nodes and links"
The article should mention Joscha Bach—because he is doing similar work and works with Open Cog and
Jeffrey
EFTA01084197
Much of Jeffrey Epstein's motivation and his foundation are to see what artificial intelligence can reveal
about human intelligence and not just the brain but cellular signal intelligence. "Signal intelligence is at
the core of understanding how diseases such as cancer are transmitted and therefore, at the core of
developing effective treatment."
As we get closer to mapping the vast mechanics of the human mind, it's possible that we'll discover that
we are more pre-determined than we think. It's possible that free will, though weighing a million
different neural filaments, is set in genetic stone—but it's also possible that the mind, in its mechanism,
can continue to push the evolutionary envelope.
EFTA01084198
Entities
0 total entities mentioned
No entities found in this document
Document Metadata
- Document ID
- 2c7120bf-690e-4bd8-8618-66080771f245
- Storage Key
- dataset_9/EFTA01084196.pdf
- Content Hash
- 57dd8eea8b4562675ec95b38f995fc27
- Created
- Feb 3, 2026