The Intersection of AI and Blockchain

The Convergence

Why This Time is Different

Benefits

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

--

--

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store