Lidiia Moor/Getty Images
- Blockchain technology could be a safeguard that prevents artificial intelligence from running amok, Coinbase said.
- Few crypto projects are currently working in AI, but there's a wide range of problems blockchain can address.
- Blockchain could help combat the spread of misinformation and shed light on AI decision-making.
The pace of development of artificial intelligence has been breathtaking, but with those strides in generative AI arise a myriad of problems experts say might only get worse as the technology becomes more sophisticated.
But there might be a guardrail that can be easily erected to prevent AI from running wild and turning our nightmares of robot overlords into reality. That necessary protective layer might be the blockchain.
That's according to Coinbase's head of research, David Duong, who writes in a new report that blockchain technology could address problems already cropping up in the booming AI space, while dealing with these issues could be a new frontier for crypto entrepreneurs.
"As applications within AI and blockchain mature, the disruptions these technologies represent may lead to areas of collaboration and the emergence of new use cases for crypto to help address specific societal challenges posed by AI," Duong wrote.
He lists five problems that have emerged with the advent of large language models that power chatbots like ChatGPT, along with solutions for each that blockchain and crypto might be able to address:
1. Demand for data resources – as generative AI relies on huge general databases to power its large language models, "A system of token incentives and penalties could ensure that the data meets the sufficient standards of quality needed by these models."
2. Demand for computational resources – Generative AI requires a lot of computing power, and blockchain could help by creating a system where resources are shared. "Decentralized networks of computing power can reward users who volunteer their idle GPUs to help relieve some of this computational burden."
3. Misinformation and disinformation – Fears of AI-driven misinformation campaigns have risen as large language models have the potential to digest false claims and spit them out as fact to users. "Blockchain and non-fungible token (NFT) technology can help combat this by establishing the provenance of images, video, music and other media."
4. Auditability – AI decision-making can be opaque to the normal user, and its algorithms inscrutable. "Integrating blockchain technology into AI models can help make those processes more transparent and allow users to scrutinize the logic and reliability of AI-produced outcomes."
5. Democratizing AI development and use – Finally, Duong wrote that crypto can make AI more accessible by allowing people to partake in its use and development on a larger scale. "Crypto platforms can help source a more diverse set of contributions, as well as redistribute access so more users can take advantage of its benefits (e.g. healthcare advice or e-commerce)."