Effect Force Private Beta accomplishment: 200,000 tasks completed in less than 48 hours.
After launching the private beta of the microtasking platform Effect Force on Thursday, Effect.AI is excited to announce that over 200,000 image classification tasks have been completed in less than two days. Given that there were only 600 workers in the private beta, we are thrilled to see the energy and enthusiasm that these early supporters have given us. While the private beta was a very small-scale version of what Effect Force will grow into, it was incredibly encouraging to see so many tasks completed. This is the first step in ultimately creating a decentralized platform that puts the potential of A.I. in the hands of everyone.
Effect Force Decentralization
Blockchain technology has a huge potential to impact the world in almost every sector, and it is about time for more blockchain applications to take off. Decentralization, one of the core ideas of blockchain technology, can be used to improve many systems, including the advancement of A.I. technology. Developing and improving A.I. algorithms often requires high quality, human annotated data. Effect Force enables this process by taking large data sets that need to be structured and labelled, and sends these tasks to a global, decentralized workforce.
With Effect Force, our aim is to make this decentralized platform accessible to everyone. Effect Force can be used to alleviate poverty by providing people with an alternative way to make a living, as workers are paid in EFX tokens in exchange for performing tasks. The successful release of our private beta demonstrates the capacity of a globally distributed workforce to perform important human intelligence tasks (HITs).
Effect Force Private Beta Results
The first stage of the Effect Force private beta consisted of over 200,000 images that over 600 of our early supporters worked to classify. This first batch of microtasks asked workers to simply classify an image as a ‘selfie’, or ‘not a selfie’. This information is then sent back to our partner, Lobster.Media, who collects and organizes billions of images. Lobster.Media needs this human touch to train their A.I. algorithms to more accurately distinguish between selfies and non-selfies.
The workers were limited to submitting 500 tasks per day in order to make sure all of our private beta users got a chance to complete some microtasks. Despite this limit, we were amazed to see how quickly tasks were being completed. The sheer volume of tasks completed in such a short time completely exceeded our expectations and showcases the strength of our workers and the capacity of Effect Force. Furthermore, our workers provided us with useful feedback about Effect Force.
Our development team is already hard at work implementing these improvements.
The Effect Force Private Beta launch was a great success and we were excited to see the overwhelmingly positive reactions in our community. The workers executed the tasks extremely quickly and efficiently, and no major problems were encountered. The data we have gathered has been invaluable and we want to express our gratitude towards the first group of workers who participated in testing the platform. These 600 workers collectively earned more than 20 000 EFX for performing these tasks, and interestingly, the resulting transactions managed to max-out the NEO blockchain at launch.
It’s really important to us that the usability and quality of Effect Force improves with each release. That’s why we will take the next few weeks to make these improvements, before releasing more images for our beta testers to work through.
Stay up to date:
To stay up to date on the latest Effect Force developments, subscribe to our YouTube channel, join our Telegram channel and follow our social feeds. All of them feature regular updates on Effect.AI and Effect Force. You can still pre-register as an Effect Force worker right now on the Effect.AI website. Make sure to do so if you want to perform tasks during the Effect
Force Private Beta.
Web | Telegram | Twitter | Facebook | YouTube | Github | Reddit | LinkedIn | Medium