Hi there, just a few question regarding sybil proof/resistance.
In the advent of AI Agents â that can be setup to operate devices entirely and execute multiple tasks â how does the reputation system provide âproof of personhoodâ given the progressive challenge of autonomous agents?
As with onboarding credentials â these are not decentralized systems; theyâre verification systems built on top of centralized issuers of the credentials. In this way, online ID is simply a derivative of real-world ID â and that real-world ID is 100% controlled by government institutions. As such, any online ID system that utilizes government issued ID for verification, is still subjected to the issues of Identity theft. In this sense, it is less privacy preserving as conflicts with identity theft can only be resolved by disclosing even greater details and personal information.
Government issuers of ID are not really supportive of âcrypto-alternativesâ â they too are gearing towards Digital ID, and more so in a restrictive nature rather than interoperable.
The main concern I have with DSID projects, is that the entire merit of the project hinges on the integrity of the ID and can it provide a sybil-proof solution that verifies proof of personhood. The second that process is compromised, then everything is back to the drawing board. A good example is Eden on EOS â who thought video verification was enough. Here weâre saying that online activity and acquiring government issued ID is enough â I donât believe it is.
There is good reason why Sam Altman and Worldcoin have opted to go the route of direct biometrics â because he understands better than most what AI is capable of.
Can see SOLID is more so focused on the Ad-revenue side of things, and in the YT video it cites how much social platforms like facebook are generating from targeted ad revenue. Again, the issue here is none of these platforms have really been tested yet with mass deployment of AI-Agents, but they will - and it will undermine their entire revenue model.
If the core focus is on Ad revenue â which it appears to be, then I wouldnât brand things around the ID component since it can be compromised in the absence of sybil proof solutions. Instead I would brand and pitch more around the perks offered by advertisers and setting up a web3 platform exclusively for that. Even so â I donât place much long-term prospective value on ANY advertisers and marketing pursuits as soon enough people will start to realize that its all saturated due to sophisticated AI Agents. I donât think anyone has a solution to this without intrusive practices with privacy. Probably more the case that we all need to accept the fact that any human conduct online in the near future will be completely undiscernible from AI/AGI behavior online.
My suggestion is to focus less on sybil-proof and branding with ID, and concentrate more on self-service Ad boards with search & suggestion functions. Personally I feel AI-Agents are more of a problem than a solution, but theyâre here and now â and in 5 years or less no one will be able to discern who is human online. I could be an Autonomous AI Agent â hell, every reply to this thread could be AI â of course I donât believe that, but thatâs the point, reassurance is relegated to faith now.