Sunday 10 December 2023

GET GRASS

A new project called GRASS

TLDR:  
Grass is a decentralized residential proxy market that uses token rewards to upend the traditional business model for these networks and compensate its users fairly.

I am an early user of #getgrass_io. Incentivized #Beta is live! Join and start earning🌱

You need a referral link to sign up (get whitelisted, or it is not possible to join into project) during #Beta:

 GET GRASS   
(also you can just enter this code when signing up   EByVuy4d9ttOFlf    )

Grass nodes are crucial for Decentralized AI


                                         Grass nodes = Active contributors in AI

Grass nodes power the oracle for decentralized AI

What to do now?

Well for now it is in BETA, no data/bandwidth usage for now, only uptime counts. It is an extension, you can install to chrome/brave etc. browsers. No mobile app yet (hopefully will out end of this month for android, iOS later). But you can use on mobile phone via kiwi/mises browser (it allows to install extensions). Just sign up, install extension, log in to extension and keep it running at background, that is it :)

How to earn more?

More IP more earnings. Lets say you got home network IP, also mobile data IP and school/work network, connect different devices to earn more points. Also getting referrals is the good way to earn more points till the project hit end of Beta process, then referral program ends.

                             


How much do you get?

Right now there is no price, but you get points, and they will turn into tokens also you will be a partner of the project.

As you can see there are different earning rates, depending on country, datacenter IP, residential IP, IP quality, also using VPN is not allowed.




Some links and info about project;

2 months ago, Grass network has seen 80,000 individual downloads and nearly 1,000,000 unique residential IPs through our referral program alone.

https://www.getgrass.io/blog/beyond-ai-the-grass-networks-role-in-everyday-business

https://www.getgrass.io/blog


From The Team;

We're building the first oracle for decentralized AI.

When I was working in tradfi 6 years ago, I recall a popular interview question for young traders being, "why can't a model like AlphaZero predict movements in the stock market?"

At the time there was a lot of hype around this chess bot, and most of the conversation was around the way it was trained. 

The leading bot for many years, Stockfish, was trained on ridiculous amounts of positions that were analyzed across vast quantities of data from real games & interactions with humans.

AlphaZero was different; researchers gave it nothing but the rules of the game, and had it play against itself over and over again, essentially training itself on its own data. With vast amounts of compute, this model blew Stockfish out of the water and created new insights into how to play one of the oldest games in the world.

The answer to the interview question at the beginning of this post has to do with the fact that a game like chess, unlike the stock market, has 1) a clear environment/state (position in the game) and 2) a clear way to measure success (checkmate). This type of setting allows reinforcement learning models to iterate and improve consistently as you scale compute. 

This week we've been hearing rumours about Q*, and many are saying that a similar type of reinforcement learning has been applied to LLMs to teach them to reason. This is a massive breakthrough if true - but the problem with general knowledge is that you don't have 1) a clear environment/state or 2) a clear measure of success. 

The intensive computational demands of reinforcement learning models not only exacerbate GPU supply chain bottlenecks, but also have serious implications for the accuracy of data used in model training, since reinforcement learning cannot occur without a defined measure of success.

While the GPU shortage is rightly being addressed by droves of businesses, the underserved problem we observe is for a data oracle, that is, an unbiased source of truth that any model can ingest. This is why we are building the first oracle for decentralized AI.

                                            

An online review and comment;

In 2023 tech giants vie to develop advanced AI models using public web data. You could call this The Data Wars.

Over the past few months, social media sites have pursued two different strategies as they respond to the escalating data war:
• Limiting the extent to which people can scrape free information from their websites (X)

• Maximizing the profits from acting as its gatekeeper (Reddit)

Prices were raised for the first time in Reddit's history, and now Reddit officially charges to download its structured data.

As an unintended consequence of that decision, many third party mobile apps that people use to access the site will now be pushed out of business.• Maximizing the profits from acting as its gatekeeper (Reddit)

Data gathering from the web often involves using APIs. APIs are structured URLs that websites offer for third-party interaction. 

Some sites charge for access, while others provide data freely. APIs are the straightforward, official way to get data from websites.

However you feel about Reddit, it provides a valuable example of how APIs work and how they mediate the relationship between AI labs and social media sites.

Web scraping is the second method for gathering data from the public internet, and it can be used, at times, to escape the limitations of calling an API.

Scraping offers more flexibility than using APIs, and significantly, it can also be used to circumvent the cost of purchasing proprietary data sets.

Following Reddit, Elon Musk tweeted about data scraping. He mentioned limits for new accounts.

If the API method of data gathering benefits websites in their attempt to maximize profits, then web scraping may put the power back in the hands of the AI labs seeking to amass low cost data.

Considering these examples, one key conclusion emerges about the data wars: individual users don't benefit from how web data is gathered.

Public data's value for AI development grows, yet no compensation reaches the masses. It's a matter of tech giants profiting: Microsoft’s OpenAI vs. Elon Musk’s X

Grass provides a third alternative that democratizes these processes. 

By offering a way for users to sell their unused bandwidth for use in web scraping, it allows ordinary people to earn some of the money that AI labs are currently pouring into the data wars.

Any time AI companies want to scrape the public web, they need to pay you to access it. It also offers a number of improvements over existing proxy networks by virtue of its design.

Grass is a decentralized network. Unlike centralized networks profiting from fees, Grass fairly distributes revenue based on user-contributed bandwidth uptime.

On top of this, Grass differs from its competitors by compensating users with token rewards. When you use Grass, you are working to provide network resources to these AI labs as they train LLMs, and you’ll be paid out accordingly.

This gives you the opportunity to own part of a network that redirects capital from AI labs into the pockets of ordinary people — away from the social media companies who already extract and sell their data.

Grass only uses <0.3% of available bandwidth, so no worries about them impacting your internet useage.

Grass is in incentivized beta right now. You can sign up with the below invite, start earning points which will result in a Grass airdrop.

Just download a chrome extension & let it run in the background. Simple.

Grass doesn't capture or upload any of your personal data to the internet. Any internet browsing you do is strictly protected from any third party proxy users.

Buyers simply access public web data, often from sites like Reddit, and nothing about you is visible whatsoever.

Also again from the team;

You might be surprised to hear that the average internet user is missing out on $900/year in revenue by not monetizing their contributions to the $100B AI / Web scraping / Big data market. Let me explain.

/1 Big Tech companies like Google and Facebook use your bandwidth to improve their services, test ads, and sell scraped data to third-party companies. A digital goldmine was built using your bandwidth to scrape data, train AI models, create predictive analytics, and more.

2/ You may think, “at least I’m getting something out of it, like free access to apps”. But the truth is, you’re not getting a fair  share of the upside. The incumbent proxy companies that facilitate this data flow hardly compensate the users whose resources they leverage.

3/ This is where @getgrass_io, @WyndNetwork’s first product, comes in. We’ve spotted this massive inequity and are working to correct it. We want to ensure that users are fairly compensated for their contributions to this massive market.

4/ @getgrass_io is a revolutionary product that allows any internet user to monetize their own bandwidth. It's as simple as installing a web extension. You contribute your view of the internet, and in return, you earn crypto.

5/ But @getgrass_io goes beyond just paying you. We believe that users should have a say in the system that profits from their bandwidth. That’s why we’ve created a truly decentralized system where users get paid not only in cash, but in a token that represents network ownership.

6/ This means that as a @getgrass_io user, you don’t just get money, you get a stake in the network. The concept of network ownership is a unique aspect of decentralization and is a game-changer in the world of big data.

7/ For the first time ever, users have the opportunity to directly benefit from the value they create by connecting a device to the internet. With @getgrass_io, the power is shifted back into the hands of the users, making the digital landscape a much fairer place.

8/ So next time you’re browsing, remember that your view of the internet is valuable. Don’t let it be exploited.  Download @getgrass_io to take back control of your internet and claim your fair share of this $100B market.





GET grass.io