Skip Navigation
Many artificial intelligence (AI) systems have already learned how to deceive humans, even systems that have been trained to be helpful and honest.
  • "Indeed, we have already observed an AI system deceiving its evaluation. One study of simulated evolution measured the replication rate of AI agents in a test environment, and eliminated any AI variants that reproduced too quickly.10 Rather than learning to reproduce slowly as the experimenter intended, the AI agents learned to play dead: to reproduce quickly when they were not under observation and slowly when they were being evaluated." Source: AI deception: A survey of examples, risks, and potential solutions, Patterns (2024). DOI: 10.1016/j.patter.2024.100988

    As it appears, it refered to: Lehman J, Clune J, Misevic D, Adami C, Altenberg L, et al. The Surprising Creativity of Digital Evolution: A Collection of Anecdotes from the Evolutionary Computation and Artificial Life Research Communities. Artif Life. 2020 Spring;26(2):274-306. doi: 10.1162/artl_a_00319. Epub 2020 Apr 9. PMID: 32271631.

    Very interesting.

  • Many artificial intelligence (AI) systems have already learned how to deceive humans, even systems that have been trained to be helpful and honest.
  • "But generally speaking, we think AI deception arises because a deception-based strategy turned out to be the best way to perform well at the given AI's training task. Deception helps them achieve their goals."

    Sounds like something I would expect from an evolved system. If deception is the best way to win, it is not irrational for a system to choice this as a strategy.

    In one study, AI organisms in a digital simulator "played dead" in order to trick a test built to eliminate AI systems that rapidly replicate.

    Interesting. Can somebody tell me which case it is?

    As far as I understand, Park et al. did some kind of metastudy as a overview of literatur.

  • Evidence is growing that LLMs will never be the route to AGI. They are consuming exponentially increasing energy, to deliver only linear improvements in performance.
  • My question is: Imagine we would put all the data input of a certain task, eg. making a meal, into text fragments and send this "sense data"-pakets ( 1 to the AI, would the AI be able to cook if the teach the AI how to give output that controlls a robot arm?

    If the answer of this question is yes, we already have a very usefull general tool. The LLM-AI will be able to controll and observe some situations. In the case that the answer is "no", I guess, it would have interesting implications.

    1 : Remember, some part of AI are already able to tell what is on a given photo. Not 100%, but good enough for a meal maybe. In some cases, it woul task "provokant".

  • Would you happily live your life in a simulation? This prescient 1970s thought experiment suggested not
  • I suspect the idea may be that given a fully constructed context, you may be limited to however that context permits you to be instead of an independently actualized/realized person.

    Could have some unwanted implication for religious people. Or maybe not. 😉

    To be frank, I cannot make much of this line. This doesn't preclude some other can make more of it. From the point of view of a reader, it would be great if the writer of this article would put a bite more into this line.

    Although if that may be what Nozick was getting at, it’s not without its own problems, much as you highlight with their position supposing existence harboring deeper meaning and significance apart from conscious creations.

    To make a long story short: I feel agnostic about this questions. At least, to a degree.

  • New research suggests that our universe has no dark matter
  • While I would certainly like to say I understood any of this. This post has not met any rigorous standard of debunking the researchers findings.

    Thats not what the posting claimed to be. You missunderstand. Either intentionally or just as a fact.

  • Would you happily live your life in a simulation? This prescient 1970s thought experiment suggested not
  • I'm unsure if this piece belongs in a group called "science". Its more about philosophy or arts, at least if you ask me.

    But the questions it explored about the nature of reality – and our supposed affinity to it – go back further.

    The idea of a Matrixlike state is as old as the first human thoughts.

    Would it matter to you if it wasn't "real"?

    That is not the same situation as Cypher in Matrix. Cypher wants himself to be fully ignorant about the fact that something like the Matrix even exist. He wants a "normal" life within the Matrix.

    In other words: He wants the exchange of a unpleasant reality for a lie. And this is the reason he is the true antagonist of the movie. The machines (eg. Agent Smith) are the evil ones, sure. Yet he is the one who makes a choice like Neo. Neo wants to beginn this heros journey with his new found mentor, while Cypher wants to go back in the old situation and even forces Neo and the others back.

    Even within the structur of the story, the movie makes a strong statement againt Cypher's approach.

    The second was that "we want to be a certain way, to be a certain sort of person", and we cannot truly be anything in the experience machine.

    I don't understand this one.

    Through the lack of "contact with any deeper reality," we would lose access to meaning and significance.

    That implies that in this reality lies deeper meaning and significance.

    If we assume this is true, then the inference is unavoideble. What about a state of doubt? Maybe it would still follow, maybe not.

    Hindriks says. Their goal was to test whether versions of the experience machine that kept participants more in contact with reality would be more acceptable to them. They found that respondents were significantly more willing to take an experience pill

    Any fictioal book or movie or video game is a kind of this experience pill. Therefor, we already know that people are willing to take the pill if they stay in contact to the reality and don't forgot the truth.

    My intuition tells me, the two reason we favore reality over a experience pill are:

      1. Genuinly angst of the creature. If you have just fictional experiences, you can easily become a victim of a predator. Its a evolutionary thing to search reality.
      1. When people want to climb a mountain, they actually want to climb the mountain. Hypnosis that makes them believe this is simply not what they want. It is a trick.
  • Global population to decline for several years ahead, study finds
  • When healthcare costs are going up.

    Thats a consequence of a world in which older people outnumbered younger one. The older one is, the more advanced medical aid this one need. In the current state of affaires, there is simply nothing we can do about it. It's just the old market mechanism: More demand means higher prices.

  • Discussion: What is the Future of Mining on Asteroids?
  • Metal meteoroids are mostly iron, which is cheap on Earth and of little use in space.

    As far as I understand your point, the mining in space needs a entire new infrastructure and new methodes in terms of metallurgy and all that.

  • Discussion: What is the Future of Mining on Asteroids?

    This thread is for a open discussion of the named question.

    Do you believe that miningoperation on asteroids are possible? Or not? Why?

    10
    What happen to The Remailer

    Does anyone has an idea what happend to the "Anonymous Remailer".

    Some years ago, there was an active scene of remailers in order to post anonym into the UseNet or send mails without a sender.

    As far as I know, there have even been technical solutions to problems like finding out whether someone is writing something based on traffic. I remember that there were even concepts for a kind of mailing list that worked in principle while respecting privacy.

    Has this been developed further?

    4
    "AMIE: A research AI system for diagnostic medical reasoning and conversations"

    From the Article: "The physician-patient conversation is a cornerstone of medicine, in which skilled and intentional communication drives diagnosis, management, empathy and trust. AI systems capable of such diagnostic dialogues could increase availability, accessibility, quality and consistency of care by being useful conversational partners to clinicians and patients alike"

    What do you say? Do you think its even possble? Could it rise privacy problems?

    0
    EU regulation of AI
    www.dw.com How the EU plans to regulate artificial intelligence – DW – 12/09/2023

    The EU Parliament and the member states agreed on a draft of their new AI Act. So what exactly will the landmark regulations entail?

    How the EU plans to regulate artificial intelligence – DW – 12/09/2023

    What do you think about the new EU AI Act?

    4
    The biology of aging
    www.nature.com Are your organs ageing well? The blood holds clues

    One organ in a person’s body can age faster than the rest — with implications for health and mortality.

    Are your organs ageing well? The blood holds clues

    I make a citation of the article. QUOTE: "But a new study1 that tracks proteins suggests that these changes aren’t uniform: an individual’s organs can age at different rates, and a given organ can age at a faster rate in one person than in another with the same chronological age." ENDQUOTE

    What do you think about this? Do you believe the may solved this problem with the pattern recognition of AI?

    3
    InitialsDiceBearhttps://github.com/dicebear/dicebearhttps://creativecommons.org/publicdomain/zero/1.0/„Initials” (https://github.com/dicebear/dicebear) by „DiceBear”, licensed under „CC0 1.0” (https://creativecommons.org/publicdomain/zero/1.0/)EN
    Endward23 @futurology.today
    Posts 5
    Comments 114