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

> But morally, its absolutely evil.

At what level of development is a human foetus anatomically distinguishable from a cow foetus?

There's no fact-based reason to draw the line in any particular place. We, humanity, don't know what "personhood" really is beyond the laws we write while guessing and the just-so stories we tell each other to justify those laws.

That's why I'm vegetarian, and why I'd become vegan quickly as soon as someone can get milk from GM bacteria. (And sell it in supermarkets).

It's also one of two reasons why I try to be nice to LLMs: just in case. (The other reason takes it as read they have no experience of existence: by being trained on humans, they'll do better and worse exactly when real humans would do better and worse, and that means worse on holiday season and when getting insulted).



> We, humanity, don't know what "personhood" really is

It's self-awareness, at least in general and as considered by a court when granting it to a chimp.

It's also why I would likely never go vegan, although I do advocate for a drastic overhaul of animal welfare standards.


Throwng in self-awareness in that context seems like a very slippery slope. That sounds like you are advocating for legal infanticide?

Besides, if you believe animals are not self-aware, why d you care about "welfare"?


I do care about welfare, and the difference between infants and zygotes is sentience.

Most medical professionals and ethicists consider 24 weeks to be the reasonable cutoff for abortion because this is when the fetus starts to develops sentience.

The reason this is relevant is because that is the first stage of development capable of having an identity relationship with the future person that fetus/infant will become.

Animals don't have to be self-ware to suffer. Not introspectively self-aware at least.


Self-awareness itself is poorly defined. So is consciousness, so is sentience, so is intelligence — and by some (but not all) definitions those are four* different things.

* or 5, if this list also has "personhood" in it


> Self-awareness itself is poorly defined.

No, not really. It has pretty standard definitions in philosophy and science, or it wouldn't have been able to be tested for over several decades. I suggest spending some time reading the wiki, it gives a pretty detailed overview.

The only point you have is about consciousness, and we don't need to understand the entire thing to understand parts of it or observe it, just like gravity.


Self awareness can mean:

• The ability to recognise one's own body as distinct from that of others, as demonstrated by plants.

• The ability to pass the mirror test, which some AI pass, but whose relevance is widely debated in animal psychology both on the grounds of sensory chauvinism and because it may cause both false positives and false negatives owing to us not being able to converse with the animals we're testing.

• Introspection, except that now we've got LLMs responding much the same way Turing hoped they would when outlining his eponymous test and suggesting that a "viva voce" interrogation would have us know if the machine was innovative or "learnt it parrot fashion"*.

As humans are also demonstrably great at confabulating reasons for their acts (see: split brain surgery, specifically experimental research on patients' cognitive functioning after surgery), it is unclear whether humans score any differently than LLMs in this test irregardless of if LLMs do or don't count as people in any other sense.

• Qualia: nobody knows.

• Mindfulness, meditation and spirituality: arguably only those who explicitly practice the appropriate mental techniques, e.g. Buddhist monks and similar.

• Public/social awareness of self-standing in community: everyone who is "cringe" fails.

* fun fact: AI critics have been stochastically parroting the stochastic parrot criticism since at least 1949


> Self awareness can mean:

Like I said, it's actually very well defined because it's been being studied for decades at this point. Just because it can sometimes be an overloaded term in colloquial usage doesn't negate that.

I again suggest you give the wiki page a read. It's quite in-depth and detailed with plenty of good references.


It's overloaded in academic research.

That makes it not well defined.

I did in fact read the Wikipedia page, and also have an A-level in philosophy, which means I've written more about this in three homework esseys than the total length of the English Wikipedia article: https://en.wikipedia.org/wiki/Self-awareness


> That makes it not well defined.

It's absolutely well defined it's just a complex topic. Most of the examples you gave in your last reply are never defined as self-awareness in an academic paper, e.g. qualia is always separate and the mirror test has always just been an indicator not the thing itself.




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

Search: