A pocket tool for Bayesian calibration. Have you ever desired external tools for the deliberate shaping of your mind? If so, I’m pleased to introduce Calibrate an open source Android app for probabilistic forecasting calibration. Do you wish that when you said something had a 75% chance, it really did...
Yesterday I discovered I had been a victim of the Typical Mind Fallacy thanks to accidentally finding out that someone very close to me doesn’t have an internal monologue, doesn’t think in words and has a visual photographic memory. Despite knowing that such people existed I thought I would know...
Epistemic status: I have made many predictions for quantitative AI development this decade, these predictions were based on what I think is solid reasoning, and extrapolations from prior data. If people do not intuitively understand the timescales of exponential functions, then multiple converging exponential functions will be even more misunderstood....
Have you ever desired external tools for the deliberate shaping of your mind?
If so, I’m pleased to introduce Calibrate an open source Android app for probabilistic forecasting calibration.
Do you wish that when you said something had a 75% chance, it really did happen about 75% of the time?
That’s what calibration means, and Calibrate helps you get there.
The App works by you first creating your predictions, about anything you like: work, play, daily life, world events, habits etc.
Then once you have made a few predictions you can track them either in list form:
Or in 3 types of graph, here is the Accuracy graph, showing your calibration. Predictions are... (read 234 more words →)
Seeing lots of criticism is discouraging, so ill just say thanks Eliezer for writing it.
Just dump the names so people have a chance of realising they are at risk then? Seems a lot better than just leaving it.
We are not in the ideal hypothetical world that can coordinate to shut down the major AI labs. So acting as if we were is not the optimum strategy. If people who see the danger start to leave the labs in protest, I suspect lab capabilities are only minimally and temporarily degraded, but the internal culture would shift further away from not killing everyone, and less real alignment work is even attempted where it is most needed.
When the inevitable comes and an existentially dangerous system is being built (which may not be obvious), I want some people in the building who can at least try and raise the alarm rather than just another yes man.
If such a strategic resignation (either individually or in groups) would ACTUALLY FOR REAL result in decent timeline increases that would be another matter.
This is weak. It seems optimised for vague non-controversiality and does not inspire confidence in me.
"We don’t expect the future to be an unqualified utopia" considering they seem to expect alignment will be solved why not?
Here is my shortlist of corrigible behaviours. I have never researched or done any thinking specifically about corrigibility before this other than a brief glance at the Arbital page sometime ago.
-Favour very high caution over realising your understanding of your goals.
-Do not act independently, defer to human operators.
-Even though bad things are happening on earth and cosmic matter is being wasted, in the short term just say so be it, take your time.
-Don’t jump ahead to what your operators will do or believe, wait for it.
-Don’t manipulate humans. Never Lie, have a strong Deontology.
-Tell operators anything about yourself they may want to or should know.
-Use Moral uncertainty, assume you are unsure about... (read more)
I guffawed when I saw Thorstads Overall ~P Doom 0.00002%, really? And some of those other probabilities weren't much better.
Calibrate people, if you haven’t done it before do it now, here’s a handy link: https://www.openphilanthropy.org/calibration
Actually per https://openai.com/blog/ai-and-efficiency/ it was AlphaZero vs AlphaGoZero.
The future of biological warfare revolves around the use of infectious agents against civilian populations.
Future? That's been the go-to biowar tactic for 3000+ years.
I had in mind a scale like 0 would be so non-vivid it didn’t exist in any degree, 100 bordering on reality (It doesn’t map to the memory question well though, and the control over your mind question could be interpreted in more than one way). Ultimately the precision isn’t high for individual estimates, the real utility comes from finding trends from many responses.
Yesterday I discovered I had been a victim of the Typical Mind Fallacy thanks to accidentally finding out that someone very close to me doesn’t have an internal monologue, doesn’t think in words and has a visual photographic memory. Despite knowing that such people existed I thought I would know if someone close to me was one. As a result I thought it would be an interesting and fun exercise to gauge the conscious perceptions of the community.
The below questions are judged on a 0-99 scale (they are not binary or predictions, the embedded Elicit formatting was just too great to not use), they attempt to find where the reader sits on... (read more)
Epistemic status: I have made many predictions for quantitative AI development this decade, these predictions were based on what I think is solid reasoning, and extrapolations from prior data.
If people do not intuitively understand the timescales of exponential functions, then multiple converging exponential functions will be even more misunderstood.
Currently there are three exponential trends acting upon AI performance, these being Algorithmic Improvements, Increasing Budgets and Hardware Improvements. I have given an overview of these trends and extrapolated a lower and upper bound for their increases out to 2030. These extrapolated increases are then combined to get the total multiplier of equivalent compute that frontier 2030 models may have over their 2020 counterparts.
Firstly...
Algorithmic... (read 1289 more words →)
I just wanted to confirm that the matching donations are managed on MIRIs end, I don't need to do anything special?