The Intersection of AI and Blockchain

10 min readApr 28, 2022


Arguably two of the most promising technologies for the decade ahead, blockchain and artificial intelligence (AI) are widely projected to play a disruptive role in practically every major industry there is — ranging from finance to healthcare, manufacturing, governance, real estate, gaming, entertainment, education, and everything in between.

It’s only natural then, that the two technologies are likely to share at least some degree of overlap in these sectors, as enterprising businesses and individuals attempt to leverage the best of both worlds to power their products and services. While both technologies are still somewhat novel and in their early stages of development, there are signs that the two are set to converge in the relatively near future.

The Convergence

Given that blockchain technology is largely used to change the way data and value are stored, used, and transferred, and artificial intelligence is used for improving efficiency, extracting utility from data, and eliminating boring and repetitive tasks, the two technologies are naturally compatible. It is no surprise then, that the two technologies are headed toward an intersection, where they will begin to enable a range of new use cases while helping to improve and augment existing ones.

There are several characteristics of blockchain technology that make it a potentially useful tool for artificial intelligence applications. For example, the decentralized nature of blockchains allows for permissionless (or permissioned) access to AI-based tools and pre-trained models, while their peer-to-peer nature can allow anybody to contribute code and help create and refine blockchain-based AIs.

With the potential to address each other’s shortcomings and pave the way for truly democratic and decentralized AIs, the two technologies are set to integrate into useful and perhaps unexpected ways.

The convergence between AI and blockchain is arguably best demonstrated by SingularityNET, a decentralized marketplace for AI algorithms. Through the platform, users can create and sell their own AI agents, which can be used for a variety of purposes, such as voice cloning, text generation, object recognition, and more. The platform is best known for its Sophia robot, which it bills as the “world’s most expressive robot” — a general-purpose robot used for research, education, entertainment, and marketing of SingularityNET and its related products.

Artificial intelligence is not only being applied to the protocol layer of many blockchains but also directly to its smart contracts. One project, known as Alethea AI, is set to use blockchain to essentially supercharge NFTs with an AI-based personality to create so-called ‘iNFTs’ — each of which acts as an interactive AI-powered avatar. Embedded with AI animations, voice synthesis, and the ability to draw on collective AI-based intelligence, these iNFTs can then be used and monetized in a variety of ways — such as being used as digital companions, helpdesk staff, user onboarding, fund managers, and more.

As evidenced by the growing number of startups and businesses that are exploring these technologies, and the record-breaking funding that AI and blockchain initiatives are receiving, it’s likely that these projects will continue to grow in scope in the months and years ahead.

Why This Time is Different

Anybody that was involved in the blockchain industry in 2017 and 2018 likely noticed the development of an array of so-called “AI-powered” blockchain platforms and applications. But while AI was a popular buzzword in 2017/18, the vast majority of projects that promised to leverage AI failed, and many were forced to pivot from the idea.

Velas, a platform that promised to use artificial intelligence to automatically optimize the efficiency of its blockchain through a unique AI-governed consensus mechanism known as “Artificial Intuition Delegated Proof-of-Stake” is one such example. The platform recently switched to a hybrid Proof-of-History and Proof-of-Stake consensus system — and removed all traces of the words “AI” and “AIDPOS” from its website.

Likewise, platforms like Thought Network, Namahe, Cerebrum, Dopamine, and dozens of other 2017/18 era blockchain + AI initiatives are either completely dead or have, as of yet, failed to deliver an AI-powered product.

The reason behind this staggering failure rate is simple. The vast majority of projects and developers failed to adequately grasp the complexity of the challenge, and the bulk of these projects was either underfunded and/or underskilled to surmount the obstacles along the way. Indeed, it is widely agreed that it costs several million dollars to train a model that can compete with the mid-tier AIs of today, whereas this was simply impossible just 3–4 years ago.

“Training GPT-3 would cost over $4.6M using a Tesla V100 cloud instance,” Chuan Li, PhD.

This, however, began to change in recent years as neural networks and deep learning algorithms became increasingly powerful as training data became larger and more accessible.

With the advent of platforms like GPT-J (an open-source alternative to GPT-3) the cost of integrating AI language models is coming down considerably, and the number of APIs being built around similar technologies is growing — allowing for easy integration. Indeed, blockchain-based projects no longer need to invent the wheel when it comes to AI, and can simply leverage off-the-shelf solutions to execute their goals.

That said, blockchain scalability still remains a challenge that could limit the integration and production of truly decentralized AIs, and at least for now, certainly prohibits the decentralized training of AI models. Even the most capable blockchains (such as Solana and NEAR) and their associated nodes are likely to be unable to efficiently train AI models. Instead, this may require the introduction of one or more application-specific AI blockchains complete with specialized deep learning nodes (e.g. those equipped with tensor processing units) and an ultra-efficient and scalable consensus system — such as sharded Proof-of-Stake (PoS) or Proof-of-Authority (PoA) to overcome.

This may take the form of an open-source, AI-oriented alternative to Dfinity’s Internet Computer blockchain, which instead of being used to serve a blockchain-based version of the internet, could be used to train and serve access to a modular decentralized AI, with each of its functions selectively siloed off as a network of smart contracts.


As we previously touched on, both blockchain and artificial intelligence are still nascent technologies that are each undergoing an evolution of their own. As a result, it can be difficult to predict exactly how each technology will mature. Nonetheless, a range of clear benefits is already being elucidated.

How AI Can Benefit Blockchain

  • Ethical governance: Right now, the blockchain industry has a governance problem. While a large number of protocols and platforms are beginning to turn to community governance via decentralized autonomous organizations and on-chain voting, the uptake rate remains low. AI could be used to carry out a set of predetermined instructions to reduce the burden of governance, and could also be used as a proxy for non-voters or as a delegate where appropriate — helping to improve the speed of decision making in decentralized organizations.
  • Dispute arbitration: AI could be used to analyze data related to blockchain-based disputes and provide recommendations to arbitrators, or in more advanced cases, the AI itself could automate parts of the arbitration process — such as identifying relevant data or issuing rulings. This may be particularly important for DAOs, which often have no way to easily identify and resolve disputes between members or with external entities.
  • Data verification: AI could be used to verify the accuracy of data entered into the blockchain, to identify patterns that may indicate fraud or other malicious activity, and to help assess the risk of certain transactions. This could help decentralized applications (particularly those considered DeFi applications) better comply with regulations without resorting to centralization.
  • Providing predictive analytics: Blockchain businesses today are typically extremely lacking on the data analytics front. Unlike traditional businesses, which often have teams dedicated to tracking key metrics, predicting future growth, and identifying opportunities, many decentralized protocols and DApps do not. By leveraging machine learning, artificial neural networks, and statistical modeling to carry out this function, blockchain-based businesses may be able to better compete with their centralized counterparts and improve their utility.
  • Improved decentralized file storage: AI can help to dramatically improve the security of decentralized file storage systems by providing advanced security features such as authentication, authorization, and encryption. It can also be used for trustless data de-duplication, load balancing, and caching while improving usability by powering a decentralized search and indexing mechanism.
  • IoT integration: AI can be used to integrate data coming from a wide variety of IoT sources, such as weather, motion, and light sensors, as well as Oracle-based data to produce high-quality data sets that can then be recorded, sold and used as needed — helping to extract usable data from the world’s more than 18 billion IoT devices.

How Blockchain Can Benefit AI

  • Secure data storage: Blockchains are widely praised for their impressive security and resilience to denial of service attacks. This has made them particularly well-suited for the storage of potentially sensitive data, as well as for highly resistant and redundant personal and business data storage. By integrating blockchain-based file storage solutions, AIs will be able to leverage tamper-proof and secure storage in their operations and benefit from the unique properties of decentralized storage networks.
  • Decentralized AI marketplaces: Right now, AI has a serious monopolization problem, since only the largest tech companies typically have the technology and resources necessary to train powerful AI models. Blockchain technology can be used to break down these barriers to entry, helping to make AI more accessible to less well-heeled enterprises and even individuals through an array of decentralized AI marketplaces.
  • Trustless AI applications: Blockchain can be used to create trustless AI applications, allowing users to leverage the AI to carry out their tasks, without needing to necessarily trust the AI provider with their data. Thanks to the advent of technologies like zero-knowledge proofs and data sharding, AIs may soon be able to access and work with only a fragment of data, which can then be combined with other fragments to produce a whole that no single entity (barring the user) can access.
  • Improved auditability: Since blockchain technology provides an immutable record of events and an easily browsable and auditable history of actions, it can also be used to systemically eliminate biases present in AI models — by allowing users to easily trace the origin of the bias and confirm it on-chain. This will help eliminate both known and as yet unknown biases in available AI models and help improve accuracy and efficiency with time.
  • Unstoppable AIs: Similar to the way that most smart contracts are essentially unstoppable and will operate until they’re either defunct or no longer used, blockchains can also host what might be the first generation of so-called unstoppable AIs — that is, AIs that cannot be directly shut down by a government or rendered offline by an attack. If this idea invokes a sense of dread in you and visions of Skynet, don’t worry, a kill switch can always be coded into the smart contract and triggered remotely.
  • Improve trust: By providing a secure, transparent, and tamper-proof way to store data and track AI algorithms, blockchain allows users to see how AIs are making their decisions, helping to ensure that they are behaving as intended. This is particularly important for AIs involved in making financial decisions — such as those managing the treasury for a DAO.

Powering New Business Models

As AIs become increasingly powerful and complex, and as their capabilities can be applied to an increased range of tasks with growing efficiency, businesses (both blockchain-based and traditional) are likely to begin leveraging them to power ever more ambitious and creative business models.

By providing data that can be used to improve the efficiency of transactions and smart contracts, as well as by providing insights that can help businesses make better decisions, AI can turn regular blockchain-enabled businesses into incredibly agile and reactive organizations that can cater to rapidly changing user demographics and regulatory environments.

This may prove particularly important for businesses operating at the bleeding edge of current regulatory frameworks, including centralized exchanges and decentralized finance protocols, which will be able to leverage AI to stay on top of regulations and automatically apply restrictions and change access to services based on various factors.

According to data from DeepDAO, there now exists well over 4,800 distinct decentralized autonomous organizations (DAOs) with a total of almost $11 billion in assets under management (AUM). The largest of these, known as BitDAO, currently manages over $2.2 billion in assets alone — putting it on par with large traditional hedge funds.

DAOs typically have no upper management structure, and often operate with a flat hierarchy, which can make it difficult to organize and execute business strategy, roadmap items, and collective decisions. AI has the potential to resolve this challenge by acting as a “superuser” of sorts, having access to specific capabilities as voted for by the DAO, allowing it to carry out and automate basic tasks, analyze and verify data to help with decision making; improve communication and collaboration among members; provide near costless user support, and assist with the management of resources.

With artificial intelligence set to operate alongside decentralized protocols, the utility gap between centralized and decentralized platforms may be set to close, helping to deliver a future where blockchain-based services are the preferred option for most people.

Want to learn more about blockchain technology and keep informed of the latest Master Ventures news? Consider following us on Twitter and Medium!

About Master Ventures

Master Ventures is a blockchain-focused venture studio helping to build the next generation of blockchain-based Web 3.0 system innovations within the crypto industry. Launched in 2018 by Founder and CEO Kyle Chassé, the company’s ethos can best be summarized in the acronym #BeBOLD: Benevolent, Open, Love, Decentralized.

Master Ventures co-creates with entrepreneurs and businesses worldwide to turn the best ideas into innovative and disruptive products. They do this by investing as strategic partners through offering advisory services to the projects they believe in. To date, Master Ventures has invested in over 40 crypto projects, including the likes of Kraken, Coinbase, Bitfinex, Reef, DAO Maker, Mantra DAO, Thorchain, and Elrond.

For any questions, please feel free to reach out to us on:

MV Website | MV Telegram| MV Twitter