The Importance of Collective Intelligence

We are at a crossroads. The proliferation of artificial intelligence has the potential to accelerate developments in technology, medicine, finance, transportation and more, but how do we build with inclusivity in mind? How do we leverage frontier technologies to create equitable access to next-gen tools?

TLDR First

• Without human intelligence, there is no artificial intelligence.
• Web2 made everyone extremely comfortable with publishing their intellectual property on the internet. Now, this content is considered “fair use” and big corporations are exploiting and monetizing this data through products like ChatGPT.
• “OpenAI was created as an open source (which is why I named it “Open” AI), non-profit company to serve as a counterweight to Google, but now it has become a closed source, maximum-profit company effectively controlled by Microsoft.” — Elon Musk via Twitter
• The revolution begins now. Decentralized Ledger Technologies provide a host of solutions to protect user privacy, make data provenance transparent and set the foundation for an inclusive data economy. Human experience, expertise and intelligence is already being used to power AI technologies. People should be rewarded for it.

Your Intelligence is the Key

If you’re reading this article, you’ve very likely had the chance to experience Chat GPT by OpenAI. I’m not going to explain it or talk it up. For the purpose of this conversation, all you need to know is that the platform is a chat bot that neatly packages information scraped from the internet into prompt-based text responses.

The model is not a creator. The model does not have ideas of its own. GPT-3’s deep learning neural network, with all of its 175 billion parameters, is made to transform internet data into content. Likewise, various media-focused AI models have been productized in the past 3 years, allowing people to generate unique artwork, music and video with simple natural language input. But what about the source material?

“You can find anything on the internet! Be careful.” — Grandma

Thanks for the quote, Grandma. Human intelligence is the foundation of great AI products and in the Web2 era, we have grown accustomed to sharing this intelligence without restriction. Once this data is published to the internet, it is widely considered to be “fair use”, allowing companies like OpenAI, Google and Amazon (and the rest of them) to leverage this freely available resource to build high-ticket products. Big corporations would have you believe this is a feature, not a bug. My problem here is not the innovation, its the monetization.

The fact is, artificial intelligence would be nothing without human knowledge and content. And for the past 30 years or so, the infrastructure to ensure inclusivity in a transparent fashion has not been readily available. Blockchain has fundamentally changed this. The companies building these incredible tools have an opportunity, and even incentive, to recognize human contribution and include data providers in their economic model. As we edge ever closer to a fully automated world, now is the time to commit to an equitable and transparent data economy. This vision is made possible thanks to decentralized ledger technologies.

It’s simple: Every time an AI model is used, payment is made on-chain and the downstream royalties flow to all contributors. This means intelligence providers, AI builders and infrastructure providers all get a slice of the pie.

Open Economy for Data and AI

Bitcoin is king. A censor-resistant, peer-to-peer system for borderless digital payments that has spawned thousands of imitators. Decentralized technologies have exploded over the past decade, giving builders the opportunity to experiment and innovate use cases in finance, art, video games, education and artificial intelligence. In many ways, founders throughout the crypto space are so fervent in their quest for a unique value proposition, they may or may not overstate the actual value add of decentralized ledger technology within their vision. When I discovered Oraichain, something just felt different.

With every update, the mission of creating an open economy for data and AI became clearer, answering every critical question: How does data come into the ecosystem and how is it stored? How does the network create equitable opportunities for data scientists and researchers? What is the long term function of Oraichain within the broader context of blockchain and what are the implications for the Web2 data economy?

You see, Oraichain was not developed specifically to take Bitcoin’s throne. And actually in terms of Layer 1 blockchains, dominance is becoming less important. We live in a multi-chain world. Oraichain fills a particular niche, an ecosystem built to service the data economy with end-to-end infrastructure, allowing for the creation of chain agnostic AI and data products that can be published to the AI Oracle and consumed by Web2 and Web3 applications alike. The major elements of this ecosystem include:

• Decentralized file storage with encryption and access controls
• AI-enabled + crowdsourced data annotation tools
• Model training and test cases
• AI + Data Marketplace
• Publish models to AI Oracle
• AI executors subnetwork
• ZK engine and hybrid rollup
• Oraichain Royalty Protocol

In essence, this is an overview of the entire data pipeline: raw data enters the ecosystem and can be added to datasets, these datasets can be structured for use in AI models, AI models are trained, tested and published to Oraichain’s AI Oracle, APIs can be integrated into smart contracts (or Web2 applications) and proof of AI execution is published on-chain. The key here is Oraichain’s Royalty Protocol; when an AI service is paid for, royalties trickle down.

We can look at aiRight on Oraichain for a simple example of how this works. Developers of an AI art generator can assign a royalty when the model is published to aiRight. At the time of generation, the “creator” can assign an additional royalty. Now every time this artwork is sold on the secondary market, royalties are distributed to both the “creator” and the AI model provider.

A more recent example is Main Street, a collective intelligence platform by Oraichain Labs US. This platform takes it a step further, recognizing the intelligence provider as contributing to the fundamental value of data + AI products. With Main Street, every data point contributed can earn a royalty token. This data is used to create products. When revenue comes in, again, royalties trickle down to the AI developer, but also continues all the way down to the data providers for their sweat equity contribution to product development.

With Main Street, the final product is an AI-enabled tool that transforms crowdsourced insights into collective intelligence, allowing users to look past the objectivity of existing AI models to discover the answers people really need on a hyper local level.

Is it safe to run at night in this neighborhood? Is it easy to find parking in the summer? What is the best area to live in? Sorry, but ChatGPT can’t help you.

Thoughts on Collective Intelligence

In general terms, collective intelligence is the product of collaborative or competitive efforts that lead to some level of consensus. This concept requires a somewhat free exchange of ideas, elevating the wisdom of the entire group in order to arrive at the best possible outcome. By contributing individual knowledge and skills, groups of people work together to solve problems.

It seems obvious, but the internet itself can be used to facilitate this on a massive scale, without institutional, economic or regional borders. Moreover, the ability to memorialize contributions to collective intelligence over time could actually allow humans to develop faster and further with every generation. Personally, I believe this potential is yet unrealized, but AI tools like ChatGPT represent a solid step towards the goal, leaving search engines with their siloed knowledge hubs behind.

Search engines are overrun by advertisers and SEO experts, making it a somewhat unreliable research tool. Ranking high in a search engine does not necessarily mean optimal relevance or infer quality knowledge. This is one reason why ChatGPT immediately took the world by storm; finally, people have free access (we’ll come back to this point) to a tool that provides direct response to their query, referencing billions of data points to produce the best possible answer. The model is great at providing objective responses, but incapable of delivering opinion. Humans need opinion, qualitative input and perspective.

ChatGPT is not a complete intelligence hub, although it may feel like one. By comparison, 430 million users flock to Reddit on a monthly basis to learn and participate in discussions, reviews and commentary about…everything. Users explore the platform to research investment opportunities, product offerings, neighborhood trends, current events and more. The value of experience is on full display here.

Now, imagine an AI tool where you can query a platform like Reddit, extracting wisdom through consensus to deliver complete context for users on a subject-specific basis. And imagine that everyone who actively contributes to this knowledge base collectively share in platform revenues. Are we getting closer?

In a way, Main Street is our first attempt to achieve this. Rather than starting with general intelligence, the focus is on local insights that will be of particular value to asset owners, buyers and renters. Quality of life is an important factor when deriving value. People need more than price per square foot and the promise of appreciation. They need opinion and insider knowledge. A broker can tell you a neighborhood is trending, but what are the residents and business owners saying?

This is only the beginning of the merger between collective intelligence, artificial intelligence and blockchain. By adopting such infrastructure and agreeing to fundamental standards, we have an opportunity to build on the open source movement to expedite innovation. Share research in real time, work together to build optimal product and have proof of contribution recorded on-chain.

My utopian (and admittedly distant) vision is a world in which there is enough incentive to cooperate, that the 1% of corporations running the world are forced to take a back seat to the will of the people.

People Over Profit

“OpenAI was created as an open source (which is why I named it “Open” AI), non-profit company to serve as a counterweight to Google, but now it has become a closed source, maximum-profit company effectively controlled by Microsoft.”

— Elon Musk via Twitter

I hope you’re not surprised that all the data on earth has been crawled and turned into a “maximum-profit” product controlled by one of the biggest companies in the world. And I know, you were never expecting to get paid for contributing your knowledge, expertise and experience to the ascension of this privately held entity’s dominance in the field of AI. But here we are with a choice: Do we continue to accept the constant exploitation of our data and intelligence -or- do we demand to be included in the profits of this artificial intelligence enterprise?

We cling so tightly to ownership. Land. Intellectual property. So tightly sometimes, that we stubbornly place ownership over progress and innovation. And when it comes to data, we sign away our rights under the guise of a privacy policy by checking a little box at the bottom of the screen. We’ve never even considered the solutions we have now.

This is more than the promise of AI Oracles for every chain. This is a cultural and political revolution. Oraichain’s infrastructure is the foundation for a fair and equitable economy that respects data, knowledge and intelligence contributions. This will become a standard, a guarantee humans won’t be left behind in the age of machines. When looking towards broader time horizons, it’s clear that automation will continue to improve efficiency for nearly every sector, but these AI technologies can’t get there without us.

*All opinions are 100% my own. But yes, I work for Oraichain Labs US.