22/07/2023
Today, I asked Bard AI whether ZKP can save humans from AI. Here's what I got from the bard.
Zero-knowledge proofs (ZKPs) are a cryptographic technique that allows one party to prove to another party that they know a certain piece of information, without revealing any other information about that information. This could potentially be used to save humans from AI in a number of ways.
For example, ZKPs could be used to verify that an AI is not biased or discriminatory. An AI could be asked to prove that it does not have any discriminatory biases in its decision-making, without revealing any other information about how it makes decisions. This would help to ensure that AI systems are fair and equitable.
ZKPs could also be used to verify that an AI is not malicious. An AI could be asked to prove that it does not have any malicious intent, without revealing any other information about its goals or intentions. This would help to ensure that AI systems are not used to harm humans.
However, it is important to note that ZKPs are not a silver bullet. They can only be used to verify that an AI knows a certain piece of information. They cannot be used to verify that an AI is actually good or bad, or that it will not harm humans in the future.
Overall, ZKPs could potentially be a useful tool for helping to ensure that AI systems are safe and beneficial for humans. However, they are not a perfect solution, and they should not be seen as a guarantee that AI will never harm humans.