Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Someone who's into this kind of thing answer me this:

Say some "Music Man" type huckster comes to your town with a "Brain Uploader" device. He promises to upload your brain to the metaverse, but all the device does is train a deep learning simulation of the victim's personality shown on a screen, while administering progressively great electric shocks on the victim until he's dead.

This gives the town the illusion that the person's soul (or consciousness, or whatever) has migrated into the metaverse. How do you detect he's lying?



I don't know about the mind uploading... but the fact that device is applying electric shocks to a person's body and physically destroying them is a massive problem. Any respectable mind uploading would leave the original in-tact. A standard autopsy of the person's body would discover the site of electrocution and the damage associated with it, incriminating the "huckster" under at least "gross negligence" (the only crime for which mens rea is not necessary) up to first degree murder.


If everyone is convinced I’m not sure he is lying. What’s the difference between a perfect simulation of a kind and the mind itself?


You'd need to prove it's perfect. I can't help but think that would require the original mind itself to judge - no one really knows how an individual thinks outside of the individual.

I suppose if you assume the person to simply be outputs in response to inputs then you could probably create a near perfect simulation, but if you assume the mind has its own internal state that makes that mind distinct, and given that I observe myself as a thinking thing I hold that to be true, then it would take introspection to get closer to the truth.


How would you feel if someone created a simulation of you and then convinced your family and friends that you are no longer needed, and that your life could be terminated? Would you go willingly?


'How do you show he is lying?' maybe makes the scenario harder to digest than it needs to be: we have great difficulty with this with, e.g., huckster politicians, since we might see that what the huckster is saying cannot be true and that the huckster benefits from it and still not be able to show that the huckster knew what they were saying was false when they said it.

Let's make tboyd's scenario rather more concrete: suppose the huckster mind-uploader tilts the marketing at a particular religious group that happens to be wealthy. He uses the group's theological terms in their marketing, which is based on a notion of soul quite similar to mainstream Christianity, although the group is more transhumanism-friendly than, e.g., the Roman Catholics, and advertises the dubious product in their newsletter. You are a former software engineer who is studying to become take orders in this group and are convinced the product cnnot do what it claims. How do you go about showing this?


It was, at the least, an unnecessary medical procedure that resulted in someone's death. So you arrest him for homicide, seize the equipment and reveal the fact that it is just some kind of electric chair. It'll go to trial and the lies will be revealed, or he'll have to give a complete confession.

Otherwise I think this is a bit of a trick question. We know that we'd be able to tell the difference between the deepfake and a specific living person -- they have a lifetime of memories, a multitude of unique relationships with various people, the ability to effortlessly distinguish a chihuahua from a blueberry muffin, etc. But if the premise is that the salesman's simulation is so good that we can't tell the difference at all, then is there actually any lie to detect?


> So you arrest him for homicide, seize the equipment and reveal the fact that it is just some kind of electric chair. It'll go to trial and the lies will be revealed, or he'll have to give a complete confession.

There may be environments today in which prosecution under the current regime would not be possible. What if he was doing this in a hospital, and had full legal immunity granted similar to Pfizer?

> We know that we'd be able to tell the difference between the deepfake and a specific living person

After seeing the advanced deep learning systems OpenAI has put out there, like GPT-3 and DALL-E, I'm not so sure. No, those systems aren't perfect, but the imperfections could be hidden from view in a low-fidelity virtual environment like FB's metaverse.

> But if the premise is that the salesman's simulation is so good that we can't tell the difference at all, then is there actually any lie to detect?

For myself and many people, human lives have a special status, and they more than just an algorithm or input/output relationship.

The idea that a simulation of a person, no matter how well-trained, being morally or legally equivalent to that person, just does not compute to people like me.

Also, a liar knows full well that he is lying -- so why should other people be confused about it, once they know it?


I know he's lying by this: mind uploading is unpossible. It would have to be body uploading. Mind is the body, mind is an attribute of the organism, of the life. You are your body, there is no escape, although the body might have autonomized emergent psychoplasmic non - physical properties that survive physical death. Even if the replica made by the "Brain Uploader" conman were to be somehow conscious it would not be you, you would still just be at the pay end of that electrocution doohickey.


I think this might be false.

I imagine some time in the far future we'll have some kind of nanotechnology that will be able to replace our biological brain cells with "nano cells".

The process will be incremental. It will start with converting a single biological cell into a nano cell and continue from there (maybe exponentially). A nano cell will behave exactly like a biological cell in every way as far as i/o goes.

At a certain point your entire brain is made of these nano cells, you have a robot brain, basically. From there your consciousness can be copied, uploaded, whatever. You wouldn't be human anymore, but you'd still be you, and there wouldn't be any ship of Theseus conundrum.


I think you have a fallacy there concerning consciousness and life being transferred into it's cyborg enhancements and then as if animating them. Consciousness and life are equivalent to me, and if you would disagree with this premise you could of course reach different conclusions and be intellectually coherent. Like I said I think your fallacy concerns your consciousness = life being transferred into your cyborg enhancements in the first place. You are not your eyeglasses, right? And you are not your exocortex of the internet. Therefore you also are not your not - living brain cell prostheses although your living system would be kind of tricked into believing so. Therefore as the last vestige of living tissue dies in Major from Ghost In A Shell, she dies and is no more and there will be only a lifeless machine left. That's my perception of the matter based on my premises.

[Edit] Of course in GIAS - universe there would still be the ghost of Major left in the spirit world. But she wouldn't link with the machine, only the life, I think. I think her ghost would essentially simply be her psychoplasmic material life autonomized from it's infrastructural base the physical body, as what comes to the ugly scientific basis for the process.

And anyway, even if by some kind of wetware - extension the original life would come to animate and interface with the synthetic, then it would still be trapped in it's that body and it's copy would simply be another separate although exactly identical individual.


How exactly would you upload the consciousness of a nano-cell-converted individual without shutting those physical nano-cells down in some way?

On a larger issue, does having a non-human organ affect your legal status as a human? Are people with pig heart valves not human anymore?


Brain in a jar, that could of course work.


> How do you detect he's lying?

Ask him to use it on himself.


But who's going to operate the machine and guide people into the Great Tomorrow?


Your question isn't answerable in its current state due to some undefined terms. Can you describe what you mean by "soul"/"consciousness"?


It means whatever you want it to mean. Whatever you want to call it, the guy is a con man. How do you expose him?


Well if we're playing a game of poorly defined terms to expose them I pants them and move on.


Is pulling people's pants down your normal response to using words that aren't well-defined?


No, but it's one way to expose them.


I see what you're doing, and I like it :)

But what I mean by expose is to reveal to people that this person is a liar. What I mean by "soul" or "consciousness" is the common usage of the term. Essentially, this person would be claiming that he can move a participant's essence or frame of view (with memories, personality, etc.) from a human body to a digital one. At least, that's how I would interpret it, and I don't think the average person would be that far off.


It's undefined terms all the way down.

"Soul", "consciousness", "essence", "frame of view", "personality", "memories" are all undefined terms. The question you really want unambiguously answered is "what the essence of a person?"

For example, if you define "essence" as "the manner in which an agent interacts with the outside world", then provided you interact with people entirely via text chat, your "essence" could be uploaded (with a sufficiently advanced simulation).

If you define "essence" as the "continuity of bodily processes that keep a human alive", then no, as your digital replica is not continuing your bodily functions.

If you define "consciousness" as a "subjective" experience, then the question is unanswerable, precisely because you've left the measure of "consciousness" up to the "subject". In addition, your definition of "subject" becomes circular, because "subject X" is the agent with "consciousness X" and "consciousness X" is the experience that "subject X" goes through. So really, when you say "consciousness" is subjective, what you're doing is punting on how you tell what a "subject" is.


Right, but that's missing the forest for the trees, isn't it?

The implication of your answer is that you COULDN'T tell that the person is lying, because you'd be too lost in semantics to tell left from right. So, you fall victim to his charade.


Are you claiming that it's not possible to define consciousness?


No, I defined it already. You keep trying to show me that my definition isn't good enough, but you haven't yet established why that discussion is relevant to my main point.


Is this your definition of consciousness?

> What I mean by "soul" or "consciousness" is the common usage of the term. Essentially, this person would be claiming that he can move a participant's essence or frame of view (with memories, personality, etc.) from a human body to a digital one. At least, that's how I would interpret it, and I don't think the average person would be that far off.

If so, the issue is that you've defined the unknown term ("consciousness"), in terms of other unknown terms ("essence", "frame of view", etc.), which doesn't actually resolve the problem of grounding the word "consciousness".

For example, to some Catholics, the "essence" of a person is their soul, which is tied to the body granted to them by God. And so, the salesman in your original question is obviously lying, because by definition, their "essence" is tied to their physical body.

The only reason the question is difficult to answer is because it's a metaphysical question. There is no physical definition of consciousness, or frame of view. So whether or not your "consciousness" shifts when you use this machine entirely depends on how your metaphysical beliefs define "consciousness".


To a certain extent, if the supposed "illusion" is convincing enough it may as well be reality. If it is possible to simulate an individual so completely that it would convince that person's closest friends and relatives of its legitimacy, then in what way is it not that person?


This is the most Silicon Valley answer.


I don't work or live in the Valley and frankly you couldn't pay me enough to. Philosophical thought on things like this is thousands of years old.


The nature of reality is a very, very old discussion. See https://en.wikipedia.org/wiki/Solipsism as an example.


is the point "is he lying?"

I mean shocking a person to death I don't get, unless you assume its a `mv` and not a `cp`, but I guess thats besides the point.


It's a bit of a simplistic metaphor, but I like the way you're thinking about the problem.


Do you want to participate?

For me the answer is simple: the most important point is that the man is ending human lives. So, he should be charged with murder and it doesn't matter what else he's doing in parallel.

I can be confident about that because we (humans) have not been given knowledge of what the soul or consciousness is. If we had, then people would have figured out how to stop death long ago -- probably during the same era when those huge pyramids and earthworks went up.

No, I don't believe modern technology has progressed beyond the ancient tech that was lost (in linear terms). If it had, we would be able to repeat those feats of geo-engineering (or at least understand their function).

So, any talk about moving the soul here or there is just a scam.


if the algo was 100% perfect probably nothing. However, most/all software has bugs or shortcomings. I imagine after 50 or so questions from a loved-one or close friend you could figure out if it was real or not.


What if the AI "person" just doesn't want to talk after 10 questions or so, because they're too busy with other virtual matters?




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: