INTRODUCTION
One of the major problem of Artificial intelligence is artificial stupidity. Intelligence comes from learning, whether you’re human or machine. Systems usually have a training phase in which they "learn" to detect the right patterns and act according to their input. Once a system is fully trained, it can then go into test phase, where it is hit with more examples and we see how it performs.
Obviously, the training phase cannot cover all possible examples that a system may deal with in the real world. These systems can be fooled in ways that humans wouldn't be. For example, random dot patterns can lead a machine to “see” things that aren’t there. If we rely on AI to bring us into a new world of labour, security and efficiency, we need to ensure that the machine performs as planned, and that people can’t overpower it to use it for their own ends.
Effect.AI offers a private, distributed ecosystem for AI development and AI related services termed The Effect Network. This network is designed to provide an alternative to services like Amazon Mechanical Turk, Fiverr, OneSpace and Guru. The Effect Network will operate fully on smart contracts deployed on the NEO blockchain. The goal will be to provide most, if not all of the services necessary for a healthy and accessible artificial intelligence market.
There are three phases viz a viz:
Mechanical Turk, Smart Market, and Effect Network, the details of the mechanical turk would be described below.
MECHANICAL TURK:
There are certain tasks that the computer currently cannot perform effectively. For instance, choosing the proper dress to wear for a wedding, of course the computer might use several algorithms to decide that given the weather, you need to wear a cloth made of silk or cotton, but the computer is not able to decide based on aesthetics; that is, it is not able to tell which is prettier on a person. Such tasks that the computer cannot perform effectively are termed human intelligence tasks. How do we train the computer to perform these tasks?
Remember Alan Turin’s Definition of an intelligent machine? It goes like this
EFFECT MECHANICAL TURK (EMT)
Effect M-Turk is a private, decentralized, marketplace for work that (for now) requires human intelligence. It is based on centralized business models like Amazon Mechanical Turk, Fiverr, OneSpace and Guru.com.
It’s a worldwide workforce on demand. A key difference with these centralized models is that the Effect M-Turk is peer to peer, meaning supply and demand are connect directly and more efficiently. Those requesting to human intelligence task (HIT for short) to the completed are connected to those willing to complete the task directly and efficiently.
On the platform, people get to post job offers (requesters post HIT), then workers (people available for the job) respond to these offers. The offers the workers see are based on certain criteria such as location, sex, age, gender etc.
Upon completion of the tasks on EMT, workers/users are paid with Effect.AI token called EFX.
There are three key points to note from the above :
Requester, Workers and HITs
Requesters:
These are people that post human intelligence task and need them to be completed. There roles on the platform is to submit task, determine the worth of a task, develop algorithms based on the response several workers to the tasks to teach their machines.HITS:
Human intelligence task; they are tasks that ordinarily the computer cannot perform perfectly well without the intervention of humans. Example of such task include capturing an actor in fear or disgust during a movie. They are the tasks requesters upload on the EMT Platform. Every task completed by a worker is more like a contract, each task should have at least the properties shown in the table below:
Workers:
They respond to HITs. They declare their interest in performing a particular task, and then, get paid based on the worth of the task.
The importance of privacy cannot be over emphasized; certain HITs might reveal personal details of the requesters; as such, requesters may make available certain details called (data credentials) that would be used for decrypting the task, before attempting to complete the task.
WHY Effects AI Mechanical Turk (EMT)?
One of the main difficulties in building intelligent machines, or in other words in knowledge engineering, is the ‘knowledge acquisition bottleneck’ – extracting knowledge from human experts.
Aside from the fact that tasks are accomplished, the main goal of EMT is to develop very intelligent machines. With the task completed, algorithms are developed, and these algorithms are taught to the machines.
Again, I quote Alan Turin on his definition of an intelligent machine:
The only way to teach machines how to perform task like humans is to allow humans performs such task first that teach the way humans perform this task to the machine. EMT makes this process a lot easier.
Artificial intelligence has evolved over the years; despite it evolving, one thing seemed to have not changed, its goal which is to develop intelligent machines, that can think cognitively, like humans.
To make the whole process more exciting, EMT incorporates blockchain technology. Upon completion of tasks, you are not paid via legal tender, but via cryptocurrency. Cryptocurrency protects user’s identity. Just imagine a currency without a government and third party like banks. A bit scary, but it is sure the future. It has several features like secured transaction, multiple transactions verified transactions etc.
EMT AND DEVELOPING COUNTRIES
One major reasons developing countries are called developing countries is because of the high rate of unemployment. Introducing EMT to these countries, would sure reduce the unemployment rate. Workers on the EMT are already employed by requesters to complete HITs. Of course this type of job can be done from home, but them completing task that provide them with source of income from the comfort of their home, would definitely reduce the rate of unemployment in the country.
MEET THE EMT TEAM
CONCLUSION
The world is changing, it is evolving, so also is the computing world. People are devising means to create super intelligent computers, not just computers that predict the outcome of an election, but computers that detect if this image contains a flower, or contain grasses. Such computers are not built with already known algorithms, but with algorithms that are gotten from the perception of humans. What better way to get this algorithm than to get them from humans, pay this humans, then teach this algorithm to the machine. It is obviously a win win situation for the developer, the machine and the humans.
This is an 
effect2018