Issue with AI alignment—diversity of opinions as competetive advantage? (as opposed to echo chambers)
I genuinely see diversity is competetive advantage—more opinions, more points of view, broader network.
EA Forum shares audience with Less Wrong and I would like to showcase circumstances that got me banned. I hope there are some lessons here.
Repurposing XKCD: https://xkcd.com/2347/
Systematic issue: gatekeeping and diversity
I think I’ve figured out something relevant and I really wanted to share my thoughts on the forum in order to facilitate feedback / comments / discussion / critique. Ended up encountering some issues getting the content published and in the process getting my account banned.
Of course I could just move on, ignore, carry on with life, find some other place. But I think that my story illustrates some systematic issue and I genuinely want to raise awareness about gatekeeping and diversity.
Gate 1
Alignment Forum FAQ: https://www.alignmentforum.org/posts/Yp2vYb4zHXEeoTkJc/welcome-and-faq
The LessWrong/Alignment Forum admins monitor activity on both sites, and if someone consistently contributes to Alignment discussions on LessWrong that get promoted to the Alignment Forum, then it’s quite possible full membership will be offered.
Gate 2
New User’s Guide to LessWrong: https://www.lesswrong.com/posts/LbbrnRvc9QwjJeics/new-user-s-guide-to-lesswrong
Gate 3
1st post—too simple
(cannot link to it: my account banned)
Gate 4
2nd post—too complex
(cannot link to it: my account banned)
Aim for a high standard if you’re contributing on the topic AI.
Gate 5
Chat with the moderator and responding to feedback, making the 2nd post simpler, reducing reading time from 10 to 7 minutes
(cannot access conversation history: banned)
Quotable soundbite:
Because it’d take too long to evaluate new posts (...) I have an overall policy of not reading it in enough detail to make the call.
Gate 6
When the chat ended I’ve sent an email to the team—I think it was polite (screenshot)
Gate 7
When did not receive any response after 3 emails—posted in the open thread—I think it was polite (screenshot)
There are frequently new “all questions welcome” AI Open Threads if that’s what you’d like to discuss.
It sounds like a lot of activities but it was once at a time, spread out in time:
message using widget
one moderator taking over, chatting via DM
after going cold: email the team
after being ignored: comment in the open thread
Freedom of Speech matters—Elon Musk and Twitter are pretty big on it. I think that on EA and LessWrong—there is no default Freedom of Speech—I think it is more like private property. Admins can be (and are) dictators but I think they are doing disservice to humanity: echo chambers, filter bubbles.
Regarding the actual content—I could be totally wrong. I could be delusional. I could be nuts. If any of these statements are true—I would like to know. It would help me navigate the consensual reality.
(nut job knowing they are crazy—can compensate for their behaviour and act accordingly—it is a nut job not knowing they are crazy that is dangerous)
(in other words—you are doing me a favour telling me that I’m nut job—however you need to explain it in a way that I’m able to understand on the level that is available to me)
For those who are curious, here is the link: https://mirror.xyz/0x315f80C7cAaCBE7Fb1c14E65A634db89A33A9637/ETK6RXnmgeNcALabcIE3k3-d-NqOHqEj8dU1_0J6cUg
I’m a human, living on Earth, my incentive is aligned with the survival.
If I didn’t care I would not engage in a discussion. But I do care and I’m realistic that certain places (such as EA and LessWrong) attract certain audience. Me posting on personal blog would not be good enough, it would not achieve primary objective which was to gather feedback.
Back to the main point: diversity as competitive advantage.
As a newcomer to the Less Wrong forum surely I was categorised as “diversity”.
It makes me wonder how many other users were rejected in a similar manner?
It makes me wonder what is the cost-benefits analysis of diversity, culture, filter bubble, echo chambers…
(I can handle criticism and rejection well, currently playing a bigger game talking about culture)
I’ll be honest.
Your thoughts in the rejected post seem pretty scattered and that makes the post very hard to follow.
And if it’s not easy to follow a post, then it’s not easy to give feedback on either.
Thank you.
“very hard to follow”—honest, genuine feedback.
That’s why when posting on my own blog I simplified and preserved the Less Wrong version as PDF as link at the bottom. I’m nicely suprised that you took the effort to read it. Now as I look at it I agree—the order of paragraphs could be better and some tangental / background / rabbit hole information removed.
All the feedback can be addressed / acted upon. If I received such feedback I would surely simplify, make some edits.
It was the “overall policy of not reading it in enough detail” that made me think about culture / diversity / echo chamber / filter bubble / confirmation bias.
I didn’t read the whole post either, just skimmed it enough it realise that it was confusing for me and that I’d probably wouldn’t really get the point you were making without reading it more than once.
Funny that you mention that.
I thought / I assumed that is the default state these days?
That’s why starting from the TLDR summary. I even explained why I use this style of writing—writing for the internet.
(the original post was in continous format, the pagination happens only when “save as PDF”)
The logic—if the summary is good enough then those interested in the content will skim it and maybe even read it. I also use headers so the table of contents is created, allowing to navigate to the relevant parts.
(from the time perspective it would be better to put the disclaimers and conflict of interest clauses towards the end, at the time I was thinking it provides a neat introduction and background)
For avoidance of the doubt—my intention is to highlight:
cultural issues
filter bubble
echo chamber
To reiterate:
initial feedback “too simple”—made it more detailed
subsequent feedback “too complex”—made it simpler
But then:
“I have an overall policy of not reading it in enough detail to make the call.”
“I concretely do not expect to approve a version of the current post as your first post.”
I guess it was a game over, but I tried anyway with posting in the open thread that got me banned.
I think I was expected to make a simpler post about somethign else to unlock my account, that would enable me to post the original thing?
Sounds overcomplicated. I didn’t have much interest in producing something random just to unlock my account, the AI alignment metric was the primary objective.
I’ve submitted the link to Hacker News (to faciliate comments) and some other AI adjacent communities because:
critical feedback
constructive criticism
meaningful discussion
crowdsourcing brainpower
figuring out fail scenarios
And until we figure out a better defintion / metric / alignment I suggest we stick to LIFE as a good starting point.
Previously I was chatting with GPT4.
To have more diverse opinions, this time I was chatting with Bard.
I would genuinely appreciate more human eyeballs and brains finding holes in what I’ve created, handy link to the blog: https://mirror.xyz/0x315f80C7cAaCBE7Fb1c14E65A634db89A33A9637/ETK6RXnmgeNcALabcIE3k3-d-NqOHqEj8dU1_0J6cUg
Bard was kind to me with praise but this is not something I was looking for. I was looking for CONSTRUCTIVE CRITICISM.
Finding holes would be better, otherwise I may accidentally think that I’ve figured something important.