How Blockchain and AI Can Work Together

Stuck between which career to pursue? Know nothing about either? No technical knowledge required for this article.

Photo by Pete Linforth from Pixabay

Buzzwords! For many, knowledge of Blockchain sounds like $$$, and knowledge of AI seems impossible. Is it realistic to try and combine the complicated technologies to create something new? Would making something that uses both require 3000+ iQ? Provided below are prompt explanations of AI, Blockchain, and then example applications using both to get your imagination rolling.

What is Blockchain?

A picture of some small square pieces with stamped letters spelling “blockchain” on a desk.
Photo by Launchpresso from Pexels

Blockchain is not as complicated as it may seem. This will be a quick introduction to it for anyone that needs to be familiarized.

5 Industry Transforming Blockchain Applications | Data Driven Investor

“Blockchain is a decentralized, distributed ledger of transactions” (Schmelzer, 2019).

Just a few key terms in Blockchain can help you understand how it works:

  • Decentralized = one company or person does not police the software themselves, for example, how Google does with Google Drive.
  • a “node” = Anyone that wants to keep a copy of all of the transactions on the blockchain and help run the chain.
  • a “ledger” = a copy of all transactions that have ever happened on-chain. A transaction can be thought of as some data being sent from one computer to another.
  • a “miner” = a computer that verifies transactions can happen by checking its transaction records (a “ledger”), for example, to make sure one person has enough money to send the amount they are trying to send. The miner gets paid in a currency such as “Bitcoin” or “Ethereum” for the computational work it did in verifying the transaction.
  • Distributed = many computers have the same file (a “ledger”) of transactions, and any of them could help verify transactions if they’d like (a “miner”) where transactions can be sent to them and verified.
  • “proof of work” = when the first computer to verify the transaction (a “miner”) sends the work they did in a file to everyone else that keeps a record of all transactions on the chain (a “ledger”). If everyone else’s copy of all transactions matches the same information as the verifier, everyone else updates their file (a “ledger”) (Bradley, 2019).
  • “smart contracts” = pieces of code that run automatically when specific coded conditions are met.

A blockchain is essentially another type of internet that anyone can connect to and use. One of the reasons it’s especially great as a financial system is because of how balances of an account are checked against numerous different copies of its transaction record, preventing a middle man from messing with the information (typically compared to a bank that would run the entire system, making their system centralized).

What is AI (Artificial Intelligence)?

Photo by Gordon Johnson from Pixabay

Understanding AI is not nearly as scary as what it is capable of. The following is enough of an explanation to be able to model how AI and Blockchain can work together.

“Artificial Intelligence (AI) is the ability of a machine or a computer program to think and learn. The concept of AI is based on the idea of building machines capable of thinking, acting, and learning like humans.” (Tunikova, 2019).

AI is considered more of a concept rather than a technology. It generally falls into 3 categories:

  • Narrow — focused on executing a single task and not thinking for itself, for example, Siri and Alexa.
  • General — has the ability to understand the context of things and make judgments based upon said things. Learns from experience and operates like a human brain, however, nothing like this exists just yet.
  • Super — intellectually superior to humans, can think for itself and operate without any human involvement, for example, Skynet.

There are other categories that make up an AI as well, such as Machine Learning and Deep Learning.

  • Machine Learning (ML) focuses on a machine’s ability to “learn” (De Jesus, 2017). ML typically is accomplished by utilizing algorithms that identify patterns and create insights from the data it is given. The algorithms are created by humans and are responsible for parsing and learning from the data. ML is used for predictions, which can help identify possible outcomes of any situation instead of a human needing to discover each one themself.
  • Deep Learning (DL) is considered the most advanced AI field and a subset of Machine Learning. DL requires a complex architecture that mimics the neural networks of a human brain, requiring huge amounts of data and colossal computing power. Some just call it an artificial neural network since it allows the machine to analyze data in a structure remarkably alike to how humans do. They don’t require a human to tell them what to do with the data — they are programmed to be fueled by the data. My favorite example would be computer vision, which basically breaks down images and videos beginning with its pixels and classifies everything the machine is “seeing” (great video here on Computer Vision from Google).

How could Blockchain and AI work together?

Endless applications of Blockchain and AI can be created; here are some ways they can work together to give you a snapshot of their compatibility.

  • A Blockchain could act as the brain of an AI — think Skynet, even though we probably should keep it from reaching that point. It’s realistic; a blockchain could be used for sharing learned data from multiple ML models around the world, allowing them all to grow off of and learn from each other’s surroundings. Nobody would own the system; no business or government could control the growth of the shared brain. It actually kind of looks like a brain when you think of all of the blocks chained together with information, acting as a decentralized intelligence… interesting to say the least.
  • AI can monitor a Blockchain — since AI can identify patterns in data, it could certainly be used to monitor a blockchain if necessary, being able to find any irregularities in the chain and helping correct them if needed.
  • Blockchain could help make Machine Learning data more accessible — for example, since a Blockchain would allow anyone to upload data and access all of that data, object recognition data could be uploaded by anyone and be accessible by any ML model greatly increasing collaboration and the speed of growing the models. Also, the data would not be owned by just one company since anyone can upload and access the data. Another fantastic example is provided by Ron Schmelzer (see Schmelzer, 2019), where he explains that “multiple smaller online businesses could share their customer personalization data in a Blockchain, allowing for better competition against companies like Amazon and Walmart who have already made their own data sets”.
  • Blockchain could improve finding errors in a Deep Learning neural network — For example, whenever some output is not as expected the specific inputted data causing this output could be easier to find inside the neural network. This is because every event on a Blockchain is recorded, making it much easier to track where something goes wrong in a neural network.
  • AI can improve mining in a Blockchain — mining requires huge amounts of energy considering the giant computations being executed. The electricity usage of mining is a huge downside to running a blockchain, but AI is already known for having the ability to greatly improve energy consumption and could certainly be used for blockchain mining (Evans, 2016).

Blockchain and AI clearly can be used to improve the functionality of each other and be used together for other applications as well. There are already plenty of companies using AI and Blockchain together for other purposes than improving the technologies themselves, here’s a list of some.

Sources

Bradley, R. (2019). Blockchain explained… in under 100 words. [online] Deloitte Switzerland. Available at: https://www2.deloitte.com/ch/en/pages/strategy-operations/articles/blockchain-explained.html [Accessed 28 Dec. 2019].

Cecille De Jesus. (2017). Artificial Intelligence: What It Is and How It Really Works. Retrieved from Futurism website: https://futurism.com/1-evergreen-making-sense-of-terms-deep-learning-machine-learning-and-ai

Evans, Rich. “DeepMind AI Reduces Energy Used for Cooling Google Data Centers by 40%.” Google, Google, 20 July 2016, blog.google/topics/environment/deepmind-ai-reduces-energy-used-for/. Accessed 31 Dec. 2019.

Oksana Tunikova. (2018, January 5). What You Need to Know About Artificial Intelligence. Retrieved December 28, 2019, from StopAd Blog: Practical content and insights, not just about ads website: https://stopad.io/blog/artificial-intelligence-facts

Schmelzer, R. (2019, October 25). AI and Blockchain: Double the Hype or Double the Value? Retrieved from https://www.forbes.com/sites/cognitiveworld/2019/10/24/ai-and-blockchain-double-the-hype-or-double-the-value/#6d58e0065eb4.

https://medium.com/media/0707f5c806284d01a4a13c7b13a91ce3/href


How Blockchain and AI Can Work Together was originally published in Data Driven Investor on Medium, where people are continuing the conversation by highlighting and responding to this story.