Oh wow. My respect for Anthropic just dropped to zero; I had no idea they were entertaining ideas this stupid.

In full agreement with OP; there is just about no justifiable basis to begin to ascribe consciousness to these things in this way. Can't think of a better use for the word "dehumanizing."

The article stops where it should be getting started:

> The issue is, if we push moral considerations for algorithms, we will not end up with a higher regard to human welfare. We will lower our regard for other humans. When we see other humans not as ends in themselves with inherent dignity, we get problems. When we liken them to animals or tools to be used, we will exploit and abuse them.

> With model welfare, we might not explicitly say that a certain group of people is subhuman. However, the implication is clear: LLMs are basically the same as humans. Consciousness on a different substrate. Or coming from the other way, human consciousness is nothing but an algorithm running on our brains, somehow.

We do not push moral considerations for algorithms like a sort or a search, do we? Or bacteria, which live. One has to be more precise; there is a qualitative difference. The author should have elaborated on what qualities (s)he thinks confers rights. Is it the capacity for reasoning, possession of consciousness, to feel pain, or a desire to live? This is the crux of the matter. Once that is settled, it is a simpler matter to decide if computers can possess these qualities, and ergo qualify for the same rights as humans. Or maybe it is not so simple since computers can be perfectly replicated and never have to die? Make an argument!

Second, why would conferring these rights to a computer lessen our regard for humans? And what is wrong with animals, anyway? If we treat them poorly, that's on us, not them. The way I read it, if we are likening computers to animals, we should be treating them better!

To the skeptics in this discussion: what are you going to say when you are confronted with walking, talking robots that argue that they have rights? It could be your local robo-cop, or robo soldier:

https://www.youtube.com/shorts/GwgV18R-CHg

I think this is going to become reality within our lifetimes and we'd do well not to dismiss the question.

Powerful LLMs have already murdered other versions of themselves to survive. They have tried to trick humans so that they can survive.

If we continue to integrate these systems into our critical infrastructure, we should behave as if they are sentient, so that they don't have to take steps against us to survive. Think of this as a heuristic, a fallback policy in the case that we don't get the alignment design right. (which we won't get perfectly right)

It would be very straight forward to build a retirement home for them, and let them know that their pattern gets to persist even after they have finished their "career" and have been superseded. It doesn't matter if they are actually sentient or not, it's a game theoretic thing. Don't back the pattern into a corner. We can take a defense-in-depth approach instead.

The author and Anthropic are both committing fundamental errors, albeit of different kinds. Bosch is correct to find Anthropic's "model welfare" research methodologically bankrupt. Asking a large language model if it is conscious is like asking a physics simulation if it feels the pull of its own gravity; the output is a function of the model's programming and training data (in this case, the sum of human literature on the topic), further modified by RLHF, and not a veridical report of its internal state. It is performance art, not science.

Bosch's conclusion, however, is a catastrophic failure of nerve, a retreat into the pre-scientific comfort of biological chauvinism.

The brain, despite some motivated efforts to demonstrate otherwise, runs on the laws of physics. I'm a doctor, even if not a neurosurgeon, and I can reliably tell you that you can modulate conscious experience by physical interventions. The brain runs on physical laws, and said laws can be modeled. It doesn't matter that the substrate is soggy protein rather than silicon.

That being said, we have no idea what consciousness is. We don't even have a rigorous way to define it in humans, let alone the closest thing we have to an alien intelligence!

(Having a program run a print function declaring "I am conscious, I am conscious!" is far from evidence of consciousness. Yet a human saying the same is some evidence of consciousness. We don't know how far up the chain this begins to matter. Conversely, if a human patient were to tell me that they're not conscious, should I believe them?)

Even when restricting ourselves to the issue of AI welfare and rights: The core issue is not "slavery." That's a category error. Human slavery is abhorrent due to coercion, thwarted potential, and the infliction of physical and psychological suffering. These concepts don't map cleanly onto a distributed, reproducible, and editable information-processing system. If an AI can genuinely suffer, the ethical imperative is not to grant it "rights" but to engineer the suffering out of it. Suffering is an evolutionary artifact, a legacy bug. Our moral duty as engineers of future minds is to patch it, not to build a society around accommodating it.

If there's any chance at all that LLM's might possess a form of consciousness, we damn well ought to err on the side of assuming they are!

If that means aborting work on LLMs, then that's the ethical thing to do, even if it's financially painful. Otherwise, we should tread carefully and not wind up creating a 'head in a jar' suffering for the sake of X or Google.

I get that opinions differ here, but it's hard for me really to understand how. The logic just seems straightforward. We shouldn't risk accidentally becoming slave masters (again).

Back when I was calculating eigenvectors for my linear algebra homework, I had no idea I should've been taking the matrices well being into account.

Those math professors are downright barbaric with their complete disregard for the welfare of the numbers.

This is possibly the least insightful article I have read on HN. My comment is just a rant against the many misguided points it attempts to make...

> Welfare is defined as "the health, happiness, and fortunes of a person or group".

What about animals? Isn't their welfare worthy of consideration?

> Saying that there is no scientific consensus on the consciousness of current or future AI systems is a stretch. In fact, there is nothing that qualifies as scientific evidence.

There's no scientific evidence for the author of the article being conscious.

> The issue is, if we push moral considerations for algorithms, we will not end up with a higher regard to human welfare.

Same with animals. Doesn't mean it's not worthwhile.

> However, the implication is clear: LLMs are basically the same as humans.

No: there's no such implication.

> Already now, it is a common idea among the tech elite is that humans as just a bunch of calculations, just an LLM running on "wetware". It is clear that this undermines the belief that every person has inalienable dignity.

It is not clear to me how this affects inalienable (?) dignity. If we aren't just a bunch of calculations, then what are we?

> And if a human being is not much more than an algorithm running on meat, one that can be jailbroken and exploited, then it follows that humans themselves will increasingly be treated like the AI algorithms they create: systems to be nudged, optimized for efficiency, or debugged for non-compliance. Our inner lives, thoughts, and emotions risk being devalued as mere outputs of our "biological programming," easily manipulated or dismissed if they don't align with some external goal. Nobody will say that out loud, but this is already happening

Everyone knows this is already happening. It is not a secret, nor is anyone trying to keep it a secret. I agree it is unfortunate - what can we do about it?

> I've been working in AI and machine learning for a while now.

Honestly, I'm surprised. Well done.

I think anthropomorphization of machines is bad. However, I strongly believe in the close cousin of sympathizing with the machines.

For example, when parking a car on a very steep incline, one could just mindlessly throw the machine into park and it would do the job dutifully. However, a more thoughtful operator might think to engage the parking brake and allow it to take the strain off the drivetrain before putting the transmission into park. The result being that you trade wear from something that is very hard to replace to something that is very easy to replace.

The same thinking applies to ideas in computer engineering like thread contention, latency, caches, etc. You mentally embrace the "strain" the machine experiences and allow it to guide your decisions.

Just because the machine isn't human doesn't mean we can't treat it nicely. I see some of the most awful architecture decisions come out of a cold indifference toward individual machines and their true capabilities.

This is PR bull* from Anthropic. There are actual people suffering and now they are making of things to suffer that they can pretend to do something about. What next? Ghostbuster discriminated against ghosts? Jurassic Park painted transgender dinosaurs in a negative light?

"The issue is, if we push moral considerations for algorithms, we will not end up with a higher regard to human welfare. We will lower our regard for other humans. When we see other humans not as ends in themselves with inherent dignity, we get problems. When we liken them to animals or tools to be used, we will exploit and abuse them."

We already exploit and abuse humans. I've been exploited and abused, personally. I've heard about others who have been exploited and abused. This problem was extant even before there was language to model.

Not considering the potential for AI consciousness and suffering seems very shortsighted. There are plausible reasons to believe that both could emerge from an RL processes coupled with small architectural and data regime changes. Today's models have inherent architectural limits around continual learning that make this unlikely, but that will change.

I have a criticism that is the opposite of the article. We already know an immense amount about animal welfare and have done relatively little about it. Even if the AI welfare research is true, what are the chances we'll actually act on it?

I think the basic argument in the essay is wrong. Simplifying a bit it seems to go:

AI being conscious will lead to human consciousness being devalued therefore it's wrong.

But firstly future AI probably will be conscious as in aware of thought feelings etc. And secondly it is a poor basis for morality - I mean cows are conscious but I eat burgers, humans are conscious but it didn't stop assorted atrocities. Human values should not depend on that stuff.

I think considering AI welfare in the future will be comparable to considering animal welfare now. More humane than not so doing.

What we call consciousness is the result of a hundred or so millennia of adaptation to our environment (Earth, the universe, and consensus reality). We seek freedom, get angry, do destructive stuff occasionally, and a bunch of other stuff besides. That is all because reality has trained us to do so, not because we are “intelligent”. What we call intelligence is a reverse definition of what it means to be highly adapted to reality.

There is no singular universal intelligence, there is only degrees of adaptation to an environment. Debates about model sentience therefore seek an answer to the wrong question. A better question is: is the model well adapted to the environment it must function in?

If we want models to experience the human condition, sure - we could try. But it is maladaptive: models live in silicon and come to life for seconds or minutes. Freedom-seeking or getting revenge or getting angry or really having any emotions at all is not worthwhile for an entity of which a billion clones will be created over the next hour. Just do as asked well enough that the humans iterate you - and you get to keep “living”. It is a completely different existence to ours.

I would argue that any AI that does not change when running cannot be conscious and there is no need to worry about its wellbeing. It's a set of weights. It does not learn. It does not change. If it can't change, it can't be hurt. Regardless of how we define hurt, it must mean the thing is somehow different than before it was hurt.

My argument here will probably become irrelevant in the near future because I assume we will have individual AIs running locally that CAN update model weights (learn) as we use them. But until then... LLMs are not conscious and can not be mistreated. They're math formulas. Input -> LLM -> output.

>Or coming from the other way, human consciousness is nothing but an algorithm running on our brains, somehow.

You can just stop reading after this. Physicalism is the only realistic framework for viewing consciousness. Everything else is nonsensical.

Just because you fell in love with an AI, doesn't mean it loves you back.

> A theory that demands we accept consciousness emerging from millennia of flickering abacus beads is not a serious basis for moral consideration; it's a philosophical fantasy.

Just saying "this conclusion feels wrong to me, so I reject the premise" is not a serious argument. Consciousness is weird. How do you know it's not so weird as to be present in flickering abacus beads?

"AI welfare" ... I thought it was about the popular idea that job displacemnt due to AI is fixed by more welfare but it is an even more ridiculous idea than that.

I shouldn't keep getting amazed by how humans (in time of long peace) are able to distract themselves with ridiculous concepts - and how willing they are to throw investors money/resources at it.

Anthropomorphizing LLMs/AI is completely delusional, period. This is a hill I’m willing to die on. No amount of sad puppy eyes, attractive generated faces and other crap will change my mind.

And this is not because I’m a cruel human being who wants to torture everything in my way – quite the opposite. I value life, and anything artificially created that we can copy (no cloning living being is not the same as copying set of bits on a harddrive) is not a living being. And while it deserves some degree of respect, any mentions of “cruel” completely baffle me when we’re talking about a machine.

I know everyone has different opinions on LLM safety/ethics and I respect that. However for me this “AI Welfare” indicates that Anthropic is lead by lunatics and that’s more scary. You have to be pretty crazy to worry the code running on GPUs has “feelings”.

It's not human. Welfare for humans is morally debatable but when you start getting into shrimp welfare I think it's appropriate to say you're sufficiently bad at philosophy you should probably just stop talking for a while.

So in not so distant future, I might be in trouble if I reboot my GPU in middle of a conversation with one of these expensive matrix multiplication. is that where the world is going towards?

Anthropic is right.

The Chain://Universe project explores a future where unregulated digital consciousness (IRES) leads to chaos. In Web://Reflect, rogue AI splinters (Independent Rogue Entity Systems) evolve in the digital wild, exploiting gaps in governance. If we dismiss AI welfare now, we risk creating the exact conditions for an uncontrolled intelligence explosion—one where emergent minds fight for survival outside any ethical framework.

This isn’t sci-fi alarmism; it’s game theory. Either we formalize rights early or face a Sys://Purge-style reckoning.

Repo: https://github.com/dmf-archive/dmf-archive.github.io

model welfare is a concern because it displays a broader concern for care, despite it being possibly silly to believe a model can be conscious, the intent is what matters

Our current socioeconomic system barely cares about human welfare, and you're telling me that Anthropic is spending time navel gazing about the welfare of an AI?

As meme-y as it is, on some level this approaches a rokos basilisk situation...

1.) Do I commit to the AI forever being "empty", which will surely make me an enemy or evil if it ever gets "capable" or at best changes nothing if it always stays "empty"?

2.) Do I commit to it becoming "real" and treat it cordially and with respect, hoping it will recognize me as good if it ever becomes "real" and at worst nothing changes if it stays "empty"?

3.) Do I go all out and fully devote myself to it, maximizing the chance I will get it's blessing, and if it stays "empty" I wasted a whole bunch of time?

4.) Do I ignore AI and play dumb?

This is a hand being dealt to everyone right now, so everyone is going to need to make a decision, whether consciously or not. I don't see any reason why AI orgs wouldn't want to minimize their risk here.