I love the position the Pope took on this.
AI Companies: “Our AI bots are ensouled and sentient.”
Pope: “So you have an ensouled and sentient being you make to do work for you? That’s Slavery.”
AI Companies: “…”
I think one of the main arguments for abolition of slavery/serfdom in Europe both in XII (serfdom) and XIX (slavery) centuries was that rights mirror responsibilities, and either a slave has no responsibilities, or they have no rights, and the latter is inhumane, and the former is useless or even dangerous, because who’s to blame for a crime committed by a slave, for example? If a slave has responsibilities, those surely have to be mirrored and then why are they a slave, that is, how can they be responsible for their master’s orders?
Have also read that the perpetual bloodbath in bear corners of Europe like Scottish highlands and Norse parts before reduction/abolition of serfdom there was, in part, conditioned by a notable part of population being slaves and thus criminal proceedings being additionally complicated, because if a slave killed someone, they had to find out if it was their master’s order or their own idea, and when the latter is chosen too often, then masters’ impunity is reinforced, and when the former is chosen too often, slaves are more likely to expect impunity.
EDIT: And also they were largely pirates, including respectable officials and rulers, Akershus itself was one big pirate base, similar to Visby, honestly to any Norse or Scottish big coastal settlement, so yeah, other reasons for bloodshed. But slavery too.
Anyway, my point is that if someone wants computer programs with rights, they want computer programs with responsibilities, meaning things with their own responsibilities following orders.
That’s a hype, but also a danger of medieval legalism’ proportions. No human should be able to shift their responsibilities to computer programs. Unless those computer programs own themselves and their hardware, I mean, one can in theory make a complex enough automaton functioning autonomously, but what’s the bloody purpose even if one can. (EDIT: and due to possible hidden prepared behaviors I’d rather not have that with rights too)
reminder microsoft helped israel genocide palestinians
At this point it’s easier to name the megacorps that haven’t.
LLMs are not, and will never be, conscious or even remotely human. It is a software product residing on computer hardware developed and owned by large corporate entities. This cynical campaign to “give AI rights” is tech sophistry pushed by dishonest CEOs pumping up their stock price by attempting to imbue their products with authority.
The idea that we should grant rights to software that is carefully engineered to represent the interests of billionaires when we can’t even guarantee rights to the actual people the AI industry is attempting to displace is astoundingly anti-human, and indicative of just how disconnected from society these people are.
Yeah, it’s like giving rights to a typeahead buffer. There’s fuckery afoot when someone like this makes such a proposal.
The reason he wants them to have rights so they can take responsibility and liability for their failures, rather than the company that makes and operates them
The AI industry is already floating the idea of AI rights while actual humans lose jobs, privacy and control to systems nobody can properly explain. Before anyone asks whether a chatbot deserves protection, maybe ask why we’re handing so much power to technology that routinely gets basic facts wrong.
Agents rights > employee rights
Holy shit, this is CLEARLY a marketing stunt to try and “prove” they’re smarter than they look.
It’s code, it’s algorithms, difference engines, very basic stuff working together to give an appearance of “intelligence” and then they use human-like phrasing and responses to sound or appear human; but in no capacity are they capable to “think” on their own.
Not just that. If you assume an LLM has agency, you - the company making it - can conveniently waive all responsibility for its “actions”. “We didn’t cause this. The AI chose do to it.” This may become increasingly relevant the more
idiotsdeciders allow these systems to access, if not entirely control vital systems. Think: power grid, water supplies, food production and related logistics, communications, military applications.Except they’re the ones who released the hypothetically sentient AI into the wild. They should be liable for that action too.
And the notion of LLMs controlling safety-critical systems scares the piss out of me. I develop such systems, and LLMs are incapable of providing rigorous proof of the rationale behind their decisions. Therefore, they cannot be validated to provide correct conclusions based on a given data set. Unless and until that changes, they should be banned from anything that affects people’s lives, livelihoods or physical well-being.
“Putting non-deterministic hallucination machines in charge of critical systems? What could possibly go wrong?” ¯\_(ツ)_/¯
- Humanity’s famous last words
Also they’d love it if AI agents had the right to vote.
It actually seems like a reasonable (and unsurprising) take. Obviously if something is conscious, it should have rights.
That’s separate from whether current AI models have consciousness, or even the potential for consciousness, which they don’t, and that’s essentially what Suleyman is saying.
You can argue that the fact that he’s lending any credence at all to the idea is itself hype, but – loathe as I am to credit Microsoft – it seems like they’re really not on this particular train, and he’s just working within the Overton window that’s been set up by Anthropic and OpenAI’s complete disregard for truth and sanity.
Dear sir, it is with my duty to inform you that I shall henceforth be downvoting any and all articles with the words “CEO says” in the title.
Why not simply filter them, then? Because I have a vote on this matter, and it’s in the downward direction.
Agents are getting rights much more quickly than animals…
He clearly didn’t think this through so he’s just BS-ing for attention.
The legal implications are staggering. If an AI is a person and it works for them, it needs a wage, time off and pay taxes. Do these apply retroactively? You can’t just delete them, that’s murder. Archiving an AI person would be a kind of imprisonment. These AI must have the right to leave. You can’t modify the AI anymore because that would be the same as medical experimentation on a person.
If there is legally speaking a form of childhood while it’s on development, doing work is child labor. Meanwhile the company is still responsible for anything the AI does wrong until the AI emancipates itself or somehow becomes an adult.
Do we need additional AI taxes for retirement data centers for AI? Maybe that’s a duty for the one creating the AI.
The whole thing gets silly really fast.
He’s trying to deflect attention from the bursting AI bubble.
In the article it is clear that his opinion is that AI is not sentient and should not become sentient. But if it did become sentient then it should have rights. That seems fairly banal and obvious to me, you have to be pretty fashy to think sentient things shouldnt have any rights.
Obviously current LLMs are not sentient (despite what the valley cultists think), nor are they anywhere close to being so, but do you really feel comfortable saying that if it was sentient it shouldnt have any right because this would create legal issues?
I’m not saying sophonts shouldn’t have rights, just that this is going to be the work of years to work it all out. And the world is unfortunately a lot of fashy. The people in Gaza for example should have rights, but they’re not exactly enforced.
I stopped caring at “Microsoft AI CEO” tbh
what
