Here I am, staring at a blank page, trying to come up with another article that will help introduce the Kimera project to new potential investors, and I’ve got writers block. What to do?
The Answer - look at the project with a fresh set of eyes.
Luckily, I have a highly intelligent wife, who, in addition to her two tertiary degrees and vast amounts of patience putting up with my idiosyncrasies, has a fantastic ability to read something and comprehend it. However, every time I mention crypto or technical computing stuff to her, I see her eyes glaze over.
Those subjects hold little interest to her. And that may also be the case for many potential investors out there:
- they may be into AI technology, but not crypto; or
- they may be into crypto, but not AI technologies; or
- they may be new to both crypto and AI technologies.
As this project encompasses and integrates both of those areas, there may be many people who read the Kimera whitepaper (WP) and don’t understand certain aspects of it.
My darling wife very graciously agreed to read through the WP and, subsequently, prepared a list of questions on facets of the project that she either didn't understand or wanted further clarification on… 3 pages of questions!
This week I’m addressing just one of the questions my wife came up with, as I believe it is important enough to deserve the attention of a stand-alone article. Expect further articles in the coming weeks that address some of the other questions on the list.
Let’s begin with the question…
(In reference to WP p.7) Kimera created and demonstrated a fully functional AGI solution back in 2010 - what has happened during the last 8 years?
Firstly, let’s remember that the research and development towards Kimera’s AGI began in 2005. It took 6 years to develop the core AGI algorithm to the point that it was successfully demonstrated in 2010. But my wife wants to know why the project (apparently) stagnated from 2010, and we are only just now seeing a public product rolled out in 2018.
There is a significant difference between (a) inventing a stand-alone prototype product and (b) rolling out that same product in a commercial environment. This is particularly true of new technological advancements, and is dangerously true of any technology that has the ability to change the very nature of life on earth as we know it, such as AGI technology.
Personally, I feel that 13 years (2005-2018) of development is a short development timeframe, particularly given the size of the project team and the novelty of the technology being developed. Other people, such as my wife, may feel this is long time, particularly the last 8 years that have elapsed since the fully functional AGI was successfully demonstrated. In the following paragraphs, I hope to show you why it hasn’t been wasted time.
To understand the nature of such progress, let us first look at a similarly impressive human advancement - flight - and think about the length of time it took to change from pioneer aviators taking solo flights to passenger aircraft performing commercial transport operations.
It was decades in the making, and there were several key factors that were addressed during that transition, not least of which was security/safety. And here’s the thing. I’m not looking to equate aircraft advancement to AGI advancement. Instead, let’s more equitably think of aircraft advancement as a parallel to computing advancement, and compare AGI advancement to spaceflight.
- The first unmanned spaceflight (Sputnik I) occurred in 1957.
- The first manned spaceflight (Vostok I) was 1961.
- The first moon orbit (Apollo 8) was 1968.
- The first moon landing (Apollo 11) was 1969.
This all occurred during the Cold War era, with two superpower countries pouring almost limitless amounts of money and resources into the Space Race - and it still took YEARS to achieve the various milestones.
Without the underlying advances in air flight, space flight never would’ve happened. The same is true for AGI - it wouldn’t be possible without the decades of previous advancements in computing.
Similarly, modern (narrow) AI research began in the 1950s, and (strong) AGI research has been around since the 1990s, with various theories and concepts being developed and tested. Let’s now turn our attention back to Kimera and look at some of the reasons why the 8 years of development since 2010 should be considered impressive, rather than lagging or slow progress.
Bear in mind, this project is not being developed by a superpower nation with unfathomably deep pockets and innumerable resources. Instead, Kimera is tech start-up with limited funds, who have been pushing the frontiers of AGI research and development, using a previously untested Cause:Effect AGI learning method.
The WP states, “…Kimera created and demonstrated a fully functional AGI solution back in 2010…”
The key words here are “fully functional”. Equate that to the first successful unmanned spaceflight. It was a proof of concept, showing it was possible and viable to continue advancements in the field. The AGI algorithm created by Kimera was shown to be functional and capable of achieving their stated goal.
It would have been irresponsible to start offering space flights to members of the public in the 1960s, because there are many other factors to consider before such a ‘discovery’ can be considered safe for mass use. Even 60 years on from humanity’s first venture into space, commercial passenger spaceflights are still out of our reach.
Likewise, it would be irresponsible to release a newly invented AGI technology into the world without undertaking due diligence and creating an ecosystem - a platform with a system of checks and balances - to ensure it can be safely and viably deployed.
Such a discovery needs to be tested, analysed, and (if necessary) improved upon. Since 2010, there have been several successful beta tests conducted with Kimera’s AGI, and the learnings from those tests have been incorporated into the evolution of the core AGI algorithm. Even during the current ICO pre-sale period, Kimera have been making improvements and upgrades to the basic “thinking” functionality of the core algorithm.
Imagine, just for a moment, that YOU created something that has the capacity to learn from it’s environment - something that has the potential to be a powerful force in shaping the future of human existence. Would you just release it into the world and see what happens? Certainly not. Worse yet, what if you developed this AGI technology and then realised the underlying technology infrastructure to safely facilitate mass adoption didn’t yet exist?
What would you do?
How would you do it?
If I was the founder of a small tech startup and I’d developed a ground-breaking AGI technology, here are some of the things I’d be considering…
How do we monetise the product?
In 2010, with cryptocurrency still in it’s infancy, the answer to global, standardised monetisation of an AGI ecosystem would have been close to impossible. Now, in 2018, with the global reach of crypto-technologies, a single, Tokenised currency built into the ecosystem is not only practical, but is also achievable.
How do we eliminate misuse of the product by various government entities and other powerful corporate forces?
Eliminating misuse of the product by powerful forces can best be achieved through the use of a decentralised network, and again, the immutability of blockchain technology that underpins cryptocurrency is a perfect fit.
How do we protect the Intellectual Property (IP)?
To protect the IP, the core algorithm needs to be installed in a system that is protected and encrypted in such a way that reverse engineering isn’t possible. While such systems have been available for some years, it was cost-prohibitive to using such a system in a decentralised multi-node ecosystem. The economics of adopting such a system is only recently viable.
How do we promote mass adoption, which is critical to the learning speed of the AGI?
The mass adoption of the technology, so critical to Kimera’s AGI success, can only be achieved through the integration of the technology into pre-existing user platforms and applications that can immediately reach millions of users. Once again, the rapid rise in popularity and use of such platforms has really only been visible on a global scale during the current decade in which we live. Subsequently, understanding that tapping into those platforms is necessary to success, could only be realised after their existence and mass adoption was recognised.
In conclusion...
I don’t believe Kimera’s AGI technology could have been released publicly in 2010. It has taken the last 8 years of testing and improving the algorithm, while simultaneously waiting for other supporting technologies to be invented and/or become globally adopted.
While my wife may not have initially seen how 8 years of development could be explained, I believe the above analysis not only explains the 8 years, but, honestly… only 8 years… wow! To have created a safe, secure, monetised ecosystem in this period of time is truly impressive.
I look forward to seeing how Kimera’s AGI will integrate into our daily lives as we move forward into the future.
If this project interests you and you’d like to find out more about it, there are two Webinars being conducted in the coming week, one is an ICO Presentation, the other is a Technical Deep-Dive into the AGI technology itself. You can register via the website: https://kimera.ai/webinars/
The Website and White Paper gives lots more info: https://kimera.ai
There are also loads of explanatory videos on their YouTube channel: https://www.youtube.com/channel/UCEvuaO0z1kqnxUIc5InHqFw
If you’re looking for other article reviews by this author, please check out this Telegram channel https://t.me/aussiesloth_ico
Proof of article ownership - ETH address: 0xcE497066921f8aE00d1128d8B9cA105f025DF210