AI: Vitalik Buterin's Radical Solution To Save Humanity

Vitalik Buterin, co-founder of Ethereum, recently suggested a “soft pause” in the use of computational resources on an industrial scale to slow down the development of potentially dangerous superintelligent artificial intelligence (AI). This proposal, along with other radical measures, aims to give humanity more time to prepare for the risks associated with such technology.

Will superintelligent AI destroy humanity?

Superintelligent AI is a theoretical model of AI that far surpasses the intelligence of the brightest humans in all areas of expertise. In a recent article, Vitalik Buterin explained that this AI could be just five years away from us, and it is impossible to predict whether its impact will be positive or negative. As a result, he proposes to reduce the computing power available globally by up to 99% for one to two years. This drastic reduction would allow for “buying time for humanity” to better understand and manage the implications of superintelligent AI.

Furthermore, he shares his thoughts on how to manage a scenario where the risk associated with AI is high. He specified that he would only advocate for a soft pause on computing hardware if he were convinced that more “muscular” measures than accountability rules were necessary. This would imply that those who use, deploy, or develop artificial intelligence could be held liable for damages caused by the model.

Proposals from Vitalik Buterin

Among the proposals for a pause in computing hardware, Vitalik Buterin suggests locating AI chips and requiring their registration. Additionally, he proposes that large-scale artificial intelligence hardware be equipped with a chip that only allows it to operate if it receives a triple signature from major international bodies once a week. These signatures would be independent of the devices and could even require proof of zero-knowledge published on a blockchain.

In summary, Vitalik Buterin’s proposal for a soft pause in the use of computing resources aims to slow down the development of artificial intelligence (AI) superintelligence. This cautious approach would allow humanity to better prepare for potential risks, thus balancing technological innovation and global security.