Where is Artificial Intelligence Headed?
Artificial Intelligence (AI) can radically improve people’s lives if we nurture its potential. AI enables machines to mimic human traits such as learning and thinking, and is able to do this at a pace and efficacy of which humans will never be capable. We can harness this power to help humanity, to enhance our own ways of thinking, to compliment our philosophical enquiry and, most importantly, to accelerate our problem solving.
AI is already disrupting every domain of society, from healthcare, transportation, finance, security, energy and education. However, the potential threat that AI poses to humanity has been at the forefront of debate surrounding the technology. SpaceX and Tesla CEO Elon Musk has notably said that AI is “our biggest existential threat”.
Some AI agents already solve complex problems using mathematical algorithms that exist inside black boxes – indecipherable to humans. Yet, while we can enable autonomous machines to have a certain ‘consciousness’, maybe it would be possible to develop AI code that is principally ethical, empowering them to perform tasks while guided by an inbuilt moral system. Of course, there are currently devices that cannot perform tasks for anything other than what they were intended. We must find a balance between controlled autonomy and independent moral reasoning; if we entirely confine AI within strict ethical boundaries, we risk hindering machine’s ability to self learn and stifle AI’s potential.
Designing machines to operate within a network that is fully open-source and decentralised is a strong strategy for maximising the beneficial impact of AI. Open-source code can be can be accessed, analysed and copied by anyone. Development of AI technology should be fully transparent, otherwise we risk losing this limitless resource of computer intelligence to one self-interested entity, such as a government or corporation. Control over this powerful technology should be held by anyone and everyone, not concentrated in the hands of a few; if we all have access to code, we drive competition and innovation, progressing together and collaboratively speeding up the evolution of AI. No one should be left behind.
Blockchain is a technique for immutably storing data in a continually growing, cryptographically secure chain of data states. Blockchain can provide a new, secure and trustworthy layer beneath AI applications. Any data hashed to a blockchain is immutable, meaning that it cannot be changed after it is written to a block. By storing data hashes associated with an AI algorithm on the blockchain, credibility and accuracy of the data can be guaranteed. Additionally, decentralisation ensures availability and access to data, preventing one single entity from gaining control over all information. Improvements and breakthroughs in AI are fundamentally dependant on data. As this data becomes increasingly important, it will determine who has control over these powerful AI algorithms of the future. Securing that data is open and accessible to all is absolutely essential to ensuring that one single entity does not gain control of the highly powerful AI algorithms of the future.
Blockchain’s immutability property offers many benefits when combined with AI. In addition to securing data through immutability, AI algorithms themselves can also be hashed and saved to the blockchain, allowing easy tracking of algorithm changes and authenticity. This representation of AI algorithms on the blockchain can be used to decentralise access to AI technology, rendering algorithms open and accessible to all. Blockchain also enables unique cryptographic token-economic models that can be incorporated into AI applications to create innovative new incentivisation systems. These crypto-economic incentives can be used to encourage preferred behaviour within AI algorithms and/or encourage contributions of data within a system.
Applying AI + Blockchain in the Real World
There are many incredibly powerful examples of the marriage of AI and blockchain in industries from energy to healthcare – it even opens the door to revolutionary concepts like self-owning cars. However, I would like to particularly focus on the Flyingcarpet.Network, an AI + Blockchain project that I founded. Flyingcarpet is a decentralised network of data-collection services executed by autonomous drones (and other IoT devices). Developers create services for the Flyingcarpet network using AI, pattern recognition and machine learning inside of a virtual AI engine (VAIE).
One use case of the network is in agriculture, where Flyingcarpet will bring innovative new economic opportunities to farmers and entrepreneurs. The network of AI-enabled data collection services will give farmers the tools they need to capitalise on the financial value of their agricultural data in the global futures markets. New economic opportunities will also be created for entrepreneurs to connect affordable IoT hardware (eg. drones) to the network and be rewarded when these devices are used to run services (eg. farmland analysis). All of this is made possible by the blockchain and artificial intelligence innovations underpinning Flyingcarpet’s architecture, coupled with a unique token-economic incentives model.
With access to the Flyingcarpet service algorithms marketplace, farmers could run countless different pre-programmed autonomous services on their IoT hardware without any software development expertise or the need for human intervention – enabling significantly more profitability through automation.
The data generated by Flyingcarpet drone services will help farmers reduce cultivation costs by providing insights into, for example, optimised use and distribution of water or fertilisers on their land. Perhaps even more significantly, Flyingcarpet will equip farmers with powerful and profitable new non-farm sources of income by enabling them to sell their farm data to global traders of agricultural futures markets, potentially earning them more from this data than they do from their entire crop.
Important Information: The views and opinions provided by City A.M.'s CryptO Insider are of those named in the article and should not be taken as investment advice. This communication is marketing material.