In this post, I propose an idea that could improve whistleblowing efficiency, thus hopefully improving AI Safety by making unsafe practices discovered marginally faster. I'm looking for feedback, ideas for improvement, and people interested in making it happen. It has been proposed before, that it's beneficial to have an efficient...
I can also unblock it manually at any point, and keep the full uncensored log of posts on a blockchain
Thanks for sharing your opinion. Regarding security: Using a full body of an email you can generate a zero knowledge using an offline tool (since all emails are hashed and signed by the email server). No new emails need to be exchanged
In this post, I propose an idea that could improve whistleblowing efficiency, thus hopefully improving AI Safety by making unsafe practices discovered marginally faster.
I'm looking for feedback, ideas for improvement, and people interested in making it happen.
It has been proposed before, that it's beneficial to have an efficient and trustworthy whistleblowing mechanism The technology that makes it possible has become easy and convenient. For example, here is Proof of Organization, built on top of ZK Email: a message board that allows people owning an email address at their company's domain to post without revealing their identity And here is an application for ring signatures using GitHub SSH keys that allows creating a... (read 413 more words →)
Regarding donor recognition, I think online recognition makes a lot of sense, e.g. colored nickname on the forum, or a badge of some sorts (like the one GWWC encourages).
Thank you for clearly laying out the arguments for donating to the Lightcone. I will!
Zero Knowledge Proofs allow proving that you are a part of a certain group without revealing who exactly. This can allow credible whistleblowing. For example, if some ML company issues a list of it's employees linked to ethereum wallets (or any other cryptographic key pair of choice), then any of it's employees would have an option to whistleblow on unsafe practices, while being 100% sure it's impossible to know who made this post.
This seem awesome to me and I wonder if anyone else shares my fascination and think this could be useful
One idea for the metric would be to elicit pairwise comparison between projects continuously, and derive ELO rating from it. The biggest advantage of this system is that it requires much less time to compare two projects rather then give some form of rating to it like: X reduction of p(Doom)
Would it be possible to make an opt-in EigenKarma layer on top of twitter (but independent from it)? I can imagine parsing say 100k most popular twitter accounts plus all of personal tweets and likes of people who opted in to the EigenKarma layer, and then building a customised twitter feed for them
What’s great about the current setup is it doesn’t produce any additional friction for users since it’s built on top of an existing post-voting system. However, one downside is that number of my given upvotes doesn’t perfectly tracks the trust I assign to the user. E.g. as pointed out by previous commenters, prolific posters could post more often and collect more likes than they intuitively “deserve.” This particular issue can be mitigated by capping the effective number of upvotes to a given user by certain defined number.
The most popular Russian speaking poker forum uses a karma system, where everyone can explicitly assign karma to each other in a range [-100:100]. It served... (read more)
Sorry, I'm not coming, decided to go to EAGx Prague
I'd love to come for a chat!
Can you elaborate on “balancing risks regarding the choppy years”?