In the months since ChatGPT was first revealed to the world, the internet has gone from jubilation and fascination to a growing list of concerns about the effect that unchecked artificial intelligence will have on society.
The blockchain ecosystem responded to the AI revolution with a surge in token prices for artificial intelligence-related crypto projects, but after the initial wave, many of those tokens have seen steep corrections as investors now look to find the substance beyond the hype.
To better understand what the future of AI and blockchain holds, Kitco Crypto had a conversation with Humayan Sheikh, the founder and CEO of Fetch.ai (FET), an artificial intelligence lab building an open, permissionless, decentralized machine learning network for the crypto economy.
According to Sheikh, there are numerous applications for AI when it comes to blockchain technology, such as AI in finance, “where it can be trained on economic data to enhance decision making and spot trends or anomalies.” Blockchain developers have also started to turn to AI to spin up smart contracts, he said.
With Fetch.ai in particular, AI has proved helpful with enhancing the projects’ Web3 tooling through the introduction of large language models into the Fetch.ai wallet, Sheikh said. “Wallet users can now enter text in the chat field to call information into the wallet. The utility and user experience can be greatly enhanced by AI and we will continue to explore in-wallet integrations.”
Future development will focus on the ability for users to transact via text input, like the ability to give a verbal command to “send eighty dollars to John.”
Other applications include the creation of AI-trained autonomous agents “that can learn and predict what useful economic work the user requires and act on those learnings with minimal or no intervention,” Sheikh noted. “AI-based Agents then become a type of personal assistant which can take the toil out of everyday life.”
As for how blockchain platforms can benefit from services like ChatGPT, one of the primary use cases noted by Sheikh is task automation. “Language models like ChatGPT can be used to understand and interpret input which can then be used in conjunction with [the] Fetch.ai tech stack to execute a response. It could be as simple as ’if’ $50 arrives in account 1 ‘then’ pay Alice. It can also be extremely complex.”
He added that the concept of how large language models and machine learning can improve products and services is being explored by a broad spectrum of projects, from blockchain-based gaming to DeFi applications and voting and governance in decentralized autonomous organizations (DAOs).
Next big advancement
When asked what he sees as the next major advancement that will come from the combination of blockchain and AI, Sheikh said that it would be the “development of autonomous systems that can operate on decentralized networks.”
“These systems could leverage blockchain’s transparency and security to make automated decisions, carry out transactions, and execute complex tasks without the need for intermediaries or centralized control,” he said. “This could lead to significant gains in efficiency, particularly in areas such as supply chain management, finance, and logistics.”
Another promising area is the integration of AI with Web3. “It has the potential to create new, decentralized applications that are more secure and transparent than traditional applications,” Sheikh noted. “By integrating AI with Web3, it would be possible to create intelligent, autonomous applications that can operate independently without the need for human intervention.”
AI-powered analytics tools that operate on blockchain networks are also a possibility. They could enable organizations to gain deeper insights into their data while maintaining the privacy and security of sensitive information.
“The combination of blockchain and AI could usher in a new era of autonomous, decentralized systems that enable greater efficiency, security, and transparency in a wide range of industries,” he said.
And to critics who say that the recent gains in AI-related tokens are all based on hype, Sheikh pushed back, noting that AI and agent technology are maturing. “We are coming out of an R&D phase and starting to reveal working technology. We are generating interest on those grounds, particularly as potential applications in the fields of supply chain management and energy distribution.”
The Fetch.ai CEO pointed to the platform’s agent technology as one of the most promising applications to date. “We believe that there is tremendous potential for this technology to transform various industries and create more efficient, decentralized systems.”
Dangers of integrating AI with blockchain
The biggest threat to platforms utilizing both AI and blockchain technology is hackers, Sheikh said. This forces projects to stay up-to-date on their due diligence with things like robust security measures and regular smart contract audits.
There are also AI-focused concerns that have started to gain attention in mainstream news, like bias and legal issues surrounding intellectual property. “These are big topics in themselves and we don’t really have the space to address them adequately, but they require careful consideration,” Sheikh said.
“Blockchain and AI are both extremely powerful technologies and taken together there is the potential to radically alter the way we conduct our lives and work,” he surmised. “AI and blockchain can interact as part of a technological apparatus that will go on to shape the future of communications.”