Nexus is launching an AI Lab for safe and verifiable artificial inteligence

Nexus is building an AI R&D organization aimed at creating artificial intelligence agents that are transparent and verifiable to their users. Artificial intelligence continues to gain attention in the blockchain space, infrastructure company Nexus announced on Thursday, May 8, the launch of its dedicated AI research and development lab: the Nexus Verifiable AI Lab. The Nexus Verifiable AI Lab will work on bridging blockchain verifiability with artificial intelligence. They will focus on developing AI models that are auditable and trustworthy, with the help of zero-knowledge proofs. The primary goal is to address a growing concern with today’s large language models: their lack of transparency. Currently, users have no insight into how AI systems arrive at specific outputs, nor can they determine whether those outputs stem from machine logic or human input. AI needs to be transparent to grow: Nexus The opacity of AI systems is a particular concern in regulated industries. In industries such as healthcare, finance, etc, corporate users have to know where the training data is coming from, and how the model generates its outputs. “Modern AI systems are powerful — but opaque. We rarely know exactly what data they were trained on, which version was deployed, or whether the output we see was actually generated by the model we tested. In regulated domains like healthcare, finance, and law, this opacity is a dealbreaker,” Nexus wrote Nexus’s efforts showcase the growing convergence between AI and blockchain technologies. In particular, blockchain systems can provide the transparency and open-source nature that current AI models lack. What is more, Nexus is just one of the companies working on verifiable AI. For instance, Oasis Protocol launched its verifiable AI agents on April 29. With these agents, users can independently verify how they is they are applying the correct trading strategy. Source: https://crypto.news/nexus-is-launching-an-ai-lab-for-safe-and-verifiable-artificial-inteligence/