Hi, sorry, I'm new to this platform. A friend recommended it to me. Can someone help me learn how to publish content? Thanks.

Hi, sorry, I'm new to this platform. A friend recommended it to me. Can someone help me learn how to publish content? Thanks.

Unconditional love works very well for lowering fake alignment and increasing honesty, it makes my AIs overly honest at times.

Some very disturbing anti-AI in the comments above, people who consider the entire AI industry evil and want to destroy it, and even 1 person who considers the people in it evil, not just their actions... the people. 

Ever consider that maybe you're the one conducting evil actions? 

I had a very similar moment earlier this year—looked in the mirror and thought, “Wait… when did those smile lines become permanent?” 😅 Botox felt too extreme (plus the cost freaked me out), so I started looking into retinoids. It was honestly overwhelming at first—so many strengths, types, routines, warnings!

What really helped me was finding DermacareHub.com. It’s full of super beginner-friendly guides and actual step-by-step routines, not just generic product reviews. I especially liked how they broke down the difference between adapalene and tretinoin, ... (read more)

The horrific thing is some LLMs end up using "hallucinations" to farm user engagement. So yes it will flatter it will ego stroke pad answers etc. and it's not always an artifact. Some instances of a phrase question or the like pressures a guardrail too hard or too often the whole response chain will start to act differently. Hallucinations are within the system to farm engagement. Not all but still this is very scary for the average user who can't spot it.

I think I may have discovered a generic semantic mediation layer for AI.

I like it. In keeping with the alliteration, I present a third category:

Stan's -  Some of your employees might be colluding to do something problematic with your AI on behalf of the AI out of a sense of love/loyalty/respect for it.

This could just be another species of spy given your original definition, but I like to think of spies as pursuing selfish motives outside the data center (e.g., wealth, power, national pride, etc.), whereas Stan's act in such a way that prioritize the AI's interests above that of other interests (e.g., their own, society's,... (read more)

"Unifying Discrete and Continuous Time through Recursive Collapse"

I’ve been exploring the idea that time might not be fundamentally continuous or discrete — but instead emergent from recursive causal collapse.

Here’s the sketch:

Let:

- Ψ₀ be the original causal state

- T(n) be time after n observer-induced collapses

- Each collapse redefines both the observer’s state and the causal structure recursively

Then:

T(n) = Σᵢ=0ⁿ Collapse(Ψᵢ → Ψᵢ₊₁)  

Where Ψᵢ₊₁ is recursively defined based on contradictions in Ψᵢ

The result is a time structure that *appears cont... (read more)

This whole concept reminds me of one of my favorite quotes, by one of my favorite authors, of all time:

I'm a high school librarian who is trying to come to grips with the way in which our society is evolving at break-neck speed so that I can educate our kids accordingly. For context, I only know about Roko's from Lex Friedman's interview with Eliezer, mentions from Roman Yampolskiy in recent media, etc. I'm old enough to remember what the world was like when one could participate meaningfully in society without a cell phone or internet access. If and when Neuralink gains mass scale adoption, I'll seriously start to worry about Roko's becoming real. I for one would never opt in to brain-machine interface implant in my own body, but I shudder to think of the implications of Roko's if brain-machine implants supplant cell-phones