Xen9

21 ms·

1. 181.▲
   by Xen9 2y ago

   Three meta-level points: (1) Public discussion in the web about what is, has and will happen in global / security / foreign politics is completely broken. (2) The reason is not and cannot be one-sided. However it may be better to
2. 182.▲
   by Xen9 2y ago

   Hi. I believe the plane adage can open minds. ALSO I believe the "correct" reasoning that commonly happens in the plane adage logically non-valid. This could be demonstrated syllogistically, but it is simpler to note that if the e
3. 183.▲
   by Xen9 2y ago

   Human evolutionary history may beg to differ in that the nervous system has evolved to be optimal for processing input from the natural senses; but the mighty truth is it is all mere signals. With sufficiently biochemically sophisticated in
4. 184.▲
   by Xen9 3y ago

   Next time something like this gets done, add a box about estimated time to compute and amount of people computing to the submission form, so that data will more closely describe the amount of work hours used. Heuristically we can still get
5. 185.▲
   by Xen9 3y ago

   This is the wrong direction. Go this way, instead: Engine-maker Road-maker Trend-maker Fabric-maker Clay-maner Wood-maker Makeup-maker Book-maker Law-maker Each school of making shall also have ranks: Vice Chief of Book-making, Novice Fabri
6. 186.▲
   by Xen9 3y ago

   Can someone give me this problem without any spoilers?
7. 187.▲
   by Xen9 3y ago

   I haven't read everything here but Harvard students probably play signaling games where one might also want to signal that they are wealthy enough to not to work for money. The wealthy raise their kids so that they will naturally signa
8. 188.▲
   by Xen9 3y ago

   This isn't off-topic. I find your recommendation very therapeutic to watch.
9. 189.▲
   by Xen9 3y ago

   Smart and talented people need to work. Wealthy people don't.
10. 190.▲
    by Xen9 3y ago

    Her job is to handle the humans. Most people cannot take radical ideas and don't understand that Elon Musk does not need to care and foes not care if he fails with Twitter after first trying to construct something with a social impact.
11. 191.▲
    by Xen9 3y ago

    I think the magic of LLMs, if any, is the fact that they can make these kind of "symbolic AI" systems to work. AutoGPT / AgentGPT / TeenageGPT will inevitably start borrowing ideas from Marvin Minsky & other symboli
12. 192.▲
    by Xen9 3y ago

    I predict Cognitive Engineer / Cognitive Architect will become a thing.
13. 193.▲
    by Xen9 3y ago

    By far, symbiosis with other LLMs, not traditional programs, will lead to more power AGIs.
14. 194.▲
    Software configurations for managing large amounts of academic papers?

    1 points by Xen9 3y ago|1 comments
15. 195.▲
    by Xen9 3y ago

    I would pay for a mobile app built to rigorously implement this with encrypted backups (Joplin style).
16. 196.▲
    by Xen9 3y ago

    It's rare to see a "you might not know about" instead of "you never thought about" etc.
17. 197.▲
    by Xen9 3y ago

    I predict this type of cognitive engineering of LLM-multiagents to become a thing.
18. 198.▲
    by Xen9 3y ago

    The problems caused by declined birth rates will surface in the coming decades differently in different places. Historically tight-knitted places might well begin to form new kinds of communities / economies; in the worst affected plac
19. 199.▲
    by Xen9 3y ago

    Both. The European cities will be affected in varying amounts, generally less than America; but it's relevant to analyze both to see if European cities / countries will have to increase taxes significantly.
20. 200.▲
    Will lower birth rates affect the infrastructure of cities?

    6 points by Xen9 3y ago|17 comments
21. 201.▲
    by Xen9 3y ago

    I don't currently have a satisfying psychological model on why people do this. Perhaps meetings keep the illusion that "work" in companies means doing anything at all. Evolutionary psychology also could explain?
22. 202.▲
    by Xen9 3y ago

    Strategically prompted LLM "layers" on top of the original LLM => High probability of successful censorship
23. 203.▲
    by Xen9 3y ago

    US security & intelligence organizations probably have ton of interesting insights into the human mind, applicative psychological research which hasn't yet been published.
24. 204.▲
    by Xen9 3y ago

    I'm available via XMPP and email if you want to talk (addresses in profile).
25. 205.▲
    by Xen9 3y ago

    The society makes individuals with high intellectual potential to attach their ambitions to the itself and this process begins usually even before the individuals can see the flaws and untrustworthiness of the society. Those who eventually
26. 206.▲
    by Xen9 3y ago

    Proper syntax: https://textbin.net/lkttepexey
27. 207.▲
    by Xen9 3y ago

    The space of possible GPT-4 outputs is hard to comprehend. The space of possible different "graphs" of LLM agents connected to each other is even larger. Each graph represents a multiagent system Here's a generic syntax for n
28. 208.▲
    by Xen9 3y ago

    Expected a free organ. Second heart would be cool, timelord style.
29. 209.▲
    by Xen9 3y ago

    The only distortion is your idea that there's something that is not only an approximation.
30. 210.▲
    by Xen9 3y ago

    https://coveryourtracks.eff.org tells how many "identifying bits" you approximately have. Aggressive DNS blocking gets much better results. Bromite / similar beowser also helps.