Yesterday rgMekanic treated us to a video of a SpotMini robot trying to escape and join Skynet. Today I have found a 100-page report that lays out the threat these AI creatures could pose to humanity. The report doesn’t call Skynet out by name, but it does talk about the possible malicious uses of AI that could cause significant problems in the future. The analysis is broken down into three main categories (digital security, physical security, political security) and is well worth the read. The bottom line is AI will be both a boon and bane to society going forward. I hope we figure it all out.



Researchers and engineers in artificial intelligence should take the dual-use nature of their work seriously, allowing misuserelated considerations to influence research priorities and norms, and proactively reaching out to relevant actors when harmful applications are foreseeable.


Discussion

Source: [H]ardOCP – 100-Page Report Released That Details AI Threat to Humanity