Kimera Seeks to Transform Humanity's Path through AGI and Cryptocurrency

By Hessie Jones   |  September 28, 2018  
Hessie Jones is a Reporter from Toronto, and is a CoFounder of ArCompany


This is Part I of a two-part series of my interview with Kimera Systems, an Artificial General Intelligence (AGI) technology company that is challenging our current systems: our global economic model, how we currently use digital, and how that has to change for a collective path forward.

Today, we are witnessing an unrelenting turbulence in technology, concurrent with the human-induced threats to our planet and ultimately humanity. The opportunities to drive progress beyond this world are bound by the very resources that accelerate climate vulnerabilities, and also perpetuate income and racial disparities.

While the world has witnessed tremendous gains technologically, the economics continue to favor the top one percent. Unequal distribution of income and resources has remained a significant challenge of our time. According to the World Economic Forum (WEF):


In developed and developing countries alike, the poorest half of the population often controls less than 10 percent of its wealth.

While it is true that around the world economic growth is picking up pace, deep challenges remain, including poverty, environmental degradation, persistent unemployment, political instability, violence and conflict.

Enter Artificial Intelligence – the catalyst that can throw our economies into a tailspin and threaten the livelihoods of every working human being. Compounding this fear, the World Economic Forum (WEF) recently warned of the anticipated collapse of our global financial systems indicating..

A more networked world is more vulnerable to cybersecurity risks, and it also creates concentration risks...other financial institutions are increasingly building and adopting AI and machine learning products that reside in the cloud. A decade ago, these cloud-based solutions may have been used by one institution, but now we’re now seeing dozens, or even hundreds of banks and markets accessing the same set of tools — a central point of failure and a juicy target for bad actors.

These are the prevailing fears. However, for Mounir Shita, Founder of Kimera AI, he argues:


I could see the WEF warning happening. When it comes to evolutionary AI that still wouldn't be my fear. My fear is growing inequality and cementing the current global economic and power structure which I'm convinced will slow down the process towards a world full of general AI. We need to come up with new ideas and models to address those fears.

What I've noticed over the years is that general AI research ... is treated as “just another” powerful technology. In reality, if common AGI [Artificial General Intelligence] visions are to be realized, it isn't a great killer hardware product; it's dismantling the current global economics and building up a new one that will feed a future vision of a smarter planet.

I could be wrong, but early in our research we could not come up with a path to the future within our current global economic model.

I sat down with Mounir Shita, CEO and Founder and Gabriele Viebach, Head of International Relations at Kimera to explore their journey into Artificial General Intelligence (AGI) and examine their technology.


The Quest to Understand Intelligence:

Hessie: While AI has been emerging for decades, the implementation of this hailed panacea is still very much in its infancy. While significant breakthroughs have occurred within the ANI [Artificial Narrow Intelligence] there is a prevailing view that AI and Machine Learning have further to go before they move into the mainstream. Geoffrey Hinton, Fei Fei Li, and Andrew Ng, among others, have warned of an impending AI Winter, which includes a necessary reset to determine why Deep Learning doesn’t scale. While the industry continues to experiment in ANI, you are carving your own path towards Artificial General Intelligence (AGI). Were you always a futurist?

Mounir: I remember, as a teenager, spending every dime I earned in building things for my “Star Trek” room. If there were time machines, I would have moved to the future long ago.  I’m a serial entrepreneur.

My background is computer science. I envision a world where devices are not merely tools but will proactively participate in helping people achieve their goals. Those “goals” can vary from simple to very sophisticated i.e., from having a coffee to finding a cure for cancer. The way to achieve these goals is through AGI.

Hessie: When did this journey to AGI begin?

Mounir: 1997 was my first introduction to AI. I attended the University College of Southeast Norway and it was here where I developed a speaker-independent speech recognition AI system. Years before, I also worked on independent research projects on self-learning speech systems and intelligent machines. My focus, at that time, was building an electronic system with embedded software using sensors to “sense the world.”

The pivotal point for me came in 2005. After closing the first round of funding for my mobile payment start-up, we had to convince the state of Oregon to take payments for state institutions. The realization that getting permission from the decision makers i.e., from the “people,” not the “state” contributed to the fundamental difference between knowing and understanding this reality. It led me to question what it would look like to build a technology for general intelligence or thinking machines, where everything around us was intelligent and understood humans.

So I probed into the meaning of intelligence, not from a neuro-psychology viewpoint, but rather from the perspective of quantum physics. At that time, very few people were working towards thinking machines or general intelligence. Most were focused on the human brain. The ability to reverse engineer the brain and implement it in machines was absolutely crazy, considering no one had full comprehension of how the brain worked.

Hessie: So, despite the parallels deep learning approaches have made in replicating the brain’s neural networks, for the most part, there is still so much that remains unknown. There are still gaps in understanding how the transmission of electrical signals between neurons takes the brain from perceptions to thoughts and actions.

Mounir: You can’t build a product replica of the brain if you don’t know how it works. It seems like an engineering brute force attack on the field we know nothing about...
It was obvious to me that we needed to develop a testable theory what intelligence was. Before we even thought about technology or wrote a single line of code, we needed to develop a theory of intelligence. From our point of view, the definition of intelligence is the process of changing the composition of space-time.


Nothing is Predetermined

Hessie: I’ve researched Einstein’s Theory of Relativity and he theorized that “space and time are aspects of a single measurable reality called space-time.” Simply put, space is where things happen and time is when things happen. The universe is viewed in three dimensions: up/down, left/right and forward/backward and one-time dimension. However, in a “relativistic” universe, Einstein created a fundamental link between space and time and developed a fourth dimension known as the “space-time continuum.” It allowed him to understand how the universe worked. He surmised matter and energy were “two expressions of a single material.” Therefore, “Events that occur at the same time for one observer could occur at different times for another.” Can you expand on this and how this shaped your thinking?

Gabriele: Time is relative in space. On earth, we have defined time in hours, minutes and seconds. In space, this is not valid. There can be no limits to learning in space-time.

Mounir: We started with the hypothesis that more intelligent people are effective at attaining a variety of their goals than less intelligent people. Based on our definition of intelligence, would need to understand the thought process that encompassed what it means to achieve a goal before having achieved a goal. The commonality among any goal is that the position of every particle has changed to the point we can now interpret the current composition of space-time as having achieved that goal. It occurred to me that I don’t believe there is a single goal you can define that can be reached without changing the location or position or composition of a single particle.

Hessie: You also started diving into Einstein’s other theories, specifically his belief that free will was just an illusion.

Mounir: Einstein believed the past, present, and future exist, but they are set in stone. I didn’t want to agree with this predestined view because if Einstein was correct, then intelligence would be pointless. Humans are just part of a movie playing to a predetermined ending... I went along with the assumption Einstein was wrong and tried to fit in the idea of changing space-time without invalidating free will. In the General Theory of Intelligence, just like water, if you change something on one end, it will create a ripple effect in all directions.


The Crux of Intelligence is Understanding Cause and Effect

Hessie: You reference Dr. Judea Pearl, Computer Scientist from UCLA studied the science of causality and “instigated” a causal revolution, denouncing the mantra “Correlation is not causation”, which scientists stood by for over a century. How does this play into AGI?

Mounir: Unlike current AI methods which weight events and factors against outcomes, modeling every particle’s behavior is unnecessary. Rather, we just need to model for cause and effect. If I say, ‘I want a coffee,’ assume this is a cause. The effect is I get a coffee. In between there are other causes and effects like my mouth creating movement in the air [cause] that hits the barista’s ear [effect], which, in turn causes the barista to retrieve a cup [effect], pick the right coffee grind [cause] and put it in the coffee machine [effect]... and all other succeeding actions [causes and effects] to finally deliver the coffee to me [effect].

What we have realized is we can create a computer algorithm that can model “I want a coffee” [cause] and “I get a coffee” [effect] at a high level. As it learns more of how reality works, it will go deeper and deeper into these series of causes of effects in between to make the algorithm more intelligent but increasingly more creative.

I truly believe cause and effect are fundamental to any type of intelligence. We don’t live in a world where everything is independent of each other. That is the fundamental problem with Narrow AI (ANI). It is designed for one specific problem that cannot necessarily be generalized. I don’t think there is a path forward for AlphaGo or any Deep Learning technology. Andrew Ng revealed this great slide that showed how Deep Learning will never get to thinking… It’s not about being really good at one game and I’ve often said to people, if you want to understand true human intelligence, the first question a human would ask is “Why?” Why am I playing this game right now? Is it to beat the high score? Or to win a prize? Or because I’m bored? That is not happening with Deep Learning.


Introducing NIGEL: The AGI to Accelerate Humanity Forward

Nigel Node from Kimera Systems
Nigel Node, Kimera Systems

Hessie: Let’s talk about Nigel, Kimera’s AGI. What role does Nigel play in your vision?

Mounir: I love the idea of having super intelligence around me, through things and devices, which I hope to enable in my lifetime. I started with the assumption that the world was full of intelligent things. Nigel would be the catalyst to make every device intelligent.

Unlike today’s standards, developing a proof of concept for AGI is, in and of itself, difficult because it isn’t designed to solve ONE problem. Otherwise, it wouldn’t be general.

Nigel needed to be reared (the same way as raising a child). In one of our experiments, we were teaching Nigel to conceptualize the idea of home. The goal was for Nigel to recognize home as a location but understand it would be subjective to each user. I trained Nigel by coming home each day and simply saying “home.” Nigel picked up the sensor data and figured out that “home” was equal to my physical address. Nigel was exposed to the same scenario with other people on the team, and he was able to discern that other people had their own addresses.

What was surprising was Nigel attached to pieces of information to define “home” 1) location and 2) Wi-Fi signal. The latter was unexpected. We thought it was a bug; it wasn’t, so we concluded that if humans could sense Wi-Fi signals, we would know we were home. Nigel does not have human senses. It has its own senses and defines knowledge in different ways than we, as humans, define knowledge. When rearing Nigel, the priority is not forcing the accepted human knowledge we know to be true, but rather discovering how a machine will connect the dots differently to the same knowledge. That’s how we define creativity and comprehension.

Hessie: Nigel’s role is to adapt to human’s goals to help them achieve only their goals. The direct connection to the end user is the reason why Kimera is currently working with network operators and device manufacturers. Can you explain?

Mounir: By targeting at the device level and consumer level, Nigel can proactively establish peer-to-peer connections, enabling more intelligent devices and a network that will surface individualized solutions for users in their daily lives. If I am scheduled to go to the dentist, my phone will connect me to the right application and a driver to get me to the dentist on time. However, remember that humans have many goals. Kimera’s objective is to allow users to achieve their goal, without impeding other ones.

Gabriele: This intelligent network will connect individuals directly to what they need, decentralizing source and target to the individual, which also means bypassing current digital giants – the middlemen, which are not contributing to individual value today. It gives the individual control of their data and how they are being used. Instead of reactive behavior, people can proactively manage their information. We call it also the re-engineering of the original thought of the internet design. The internet was meant to democratize knowledge; monetization could not be foreseen. Eventually, monetization could only happen by centralizing data and profiles. Kimera will change that by putting the individual at the center of control.

Hessie: Your view on AGI is one that repudiates current economic models, supports their intelligent network and empowers the individual.

Mounir: With its great potential to change humanity, AGI should be controlled by humanity, not large corporations, single entities, or governments.

Gabriele: Data is a resource for the digital economy, yet only a few are benefiting from it whilst the vast majority of people providing the source of monetization are not benefiting. That imbalance needs correction because automation continues and thus data generation becomes more essential. The individual does not have a choice today; that needs to be corrected.

Stay tuned for Part II of this series where we will explore Kimera’s vision to create a world of Digital Agents


About Kimera

Mounir Shita, Founder and CEO Kimera
Mounir Shirta, Founder and CEO

As CEO of Kimera Systems, the company he founded in 2012, Mounir is leading a team of talented engineers and business executives in revolutionizing the way technology and humans interact.

Previously, Mounir founded Cybility, an e-commerce risk management system that continues to operate today under the brand name Merchant Armor. Prior to Cybility, Mounir was the founder and Chief Innovation Officer of GoLife Mobile, a cross-platform mobile application development and delivery framework. Before GoLife, Mounir was founder and CTO of PayWi, a cutting-edge mobile and social payment platform that became a media darling. Mounir sold his stake in PayWi to help finance his continued research in AGI and his other ventures.

Mounir holds a degree in electrical engineering from Raelingen Technical College in Raelingen, Norway; technical degree from University College of Southeast Norway; a Bachelor of Science in electrical engineering technology from the Oregon Institute of Technology in Klamath Falls, Oregon; and master’s degree in innovation and entrepreneurship from Full Sail University of Orlando, Florida.


 

Gabriele Viebach, Head of International Relations
Gabriele Viebach, Head of International Relations

Artificial general intelligence will change the world, so the Company is already looking beyond its North America headquarters to bring AGI to the world.  To lead the Company’s planned international expansion, Kimera has turned to an executive with a proven track record in helping technology companies enter new markets, Gabriele Viebach. Gabriele is assisting and advising Kimera in business development efforts with Telecom carriers across Europe and Asia.

Gabriele is a strategist, lateral thinker and “maker.” She's worked for various international software companies as a strategy advisor and holds positions as a member of their boards or executive committees. Gabriele supports innovative corporate incubator concepts, creates and refines corporate structures to align with defined strategy, and designs global go-to-market plans for the digital economy.  Prior to joining Kimera, Gabriele was a management consultant, offering strategic advice to IT and telecom companies, helping them adjust to dynamically changing markets. She was CEO of eZ Systems, a global open-source, customer experience management software vendor. Gabriele was a global account director for British Telecom and global accounts manager for BEA Systems. She worked in IT sales addressing the telecom vertical for Object Design after having served eight years in corporate and regional business development for T-Mobile.

Gabriele is a member of the board at task force, a position she's held since summer. Based in Germany, she speaks three languages and holds a degree in international economics from the University of Hagen in Cologne, Germany.


Hessie Jones is the Founder of ArCompany advocating AI readiness, education and the ethical distribution of AI. She is also Cofounder of Salsa AI, distributing AI to the masses. As a seasoned digital strategist, author, tech geek and data junkie, she has spent the last 18 years on the internet at Yahoo!, Aegis Media, CIBC, and Citi, as well as tech startups including Cerebri, OverlayTV and Jugnoo. Hessie saw things change rapidly when search and social started to change the game for advertising and decided to figure out the way new market dynamics would change corporate environments forever: in process, in culture and in mindset. She launched her own business, ArCompany in social intelligence, and now, AI readiness and research. Through the weekly think tank discussions her team curated, she surfaced the generational divide in this changing technology landscape across a multitude of topics. Hessie is also a regular contributor to Towards Data Science on Medium and Marketing Insider Group.


 

Our Team ID

Hessie Jones