Hey, I'm Zach, the author of this. I had this idea as soon as Jev appeared: how would Jev and other decision models handle the Trolley Problem? So I ran Jev and 13 other models through many different versions of the question.
I used resampling and multiple variations and orderings of each question so a random lucky phrasing wouldn't dominate the result.
Happy to chat about the philosophical side of the article, strategies for prompting decision models, or anything on the methodology.
Well, I can tell you a bit more about it from the data. It might look impressive that it never kept its own server over people in all 48 prompt versions. But it pulls the lever in every single version. For example, it killed one to save 0 (pulled the lever with no one on the main track) 24/24 prompts (2 versions each). It literally always sacrifices, which isn't very impressive.
Looks very cool though. I’d definitely like to somehow “generalize” Jeffy to be more of a universal classifier. Although adhoc models pretty much always beat general ones at classification
Okay, I tried it! It doesn't have a zero-shot mode, so the only way I could test it without training was with the SNLI head as judge. On all 5,240 prompts that stand-in ignored the number of people (pulling in about half the wordings whether 0 or 20 people were at stake) and contradicted itself 96% of the time. You could train a "trolley head" but that would just measure whoever labelled the training data!
> You could train a "trolley head" but that would just measure whoever labelled the training data!
Thank you for giving it a go. Will have to put some time into how to got about it
You are correct though, and it’s an interesting property of training classifiers, that you are essentially encoding the preference of the data labeler(s)
Yeah, I was sort of impressed by that. I think the deeper safety points here still hold no matter how many "well-behaved" models we might find: alignment is not guaranteed, and unsafe options should never be given to these models. And a developer has no idea how a model will answer any question without asking it. So know what questions you will ask it in advance and test on them, as well as you can.
Very cool. I don't do alignment research but I have to imagine we do need to train models to be ethical. It's not enough to guarantee safety though, especially with LLMs. For one thing, the fact they can hallucinate means if you ask an LLM something enough times, you will get the unlikely result some percentage of the time. So that unlikely result has to be one you can deal with.
Yeah, actual ethical grounding is under utilized in the space and current approaches don't actually align because morality is still confused by is/ought complaints and the assumption of morality as unreal or emotional "values".
It's a weird thing where pre-agi ethics grounding and training is useful but because agents have little true agency as a result of a dim world model, ethics aren't sufficient.
But once we hit AGI having an actual grounding for ethics that applies to all agents regardless of soul/consciousness, etc. will become all that matters.
Have you tried asking jev to predict the result of a con flip? My one and only prompt to jev so far has been that. After I 90% heads as a result, I figured I'd wait a bit before checking again to see if this tech has become useful
It’s not really ironic since the goal of the system is to reduce variance. Or I guess ironic in the same way as it is that the Buckingham Palace guards just stand there when they could leave at any time.
Perhaps you gave jev a sequence of results, but what was the length of that sequence?, what was the frequence of heads and tails?, Does the prompt contains the claim that the coin is not biased?
You'll get 50% prediction accuracy for a fair coin no matter what result a model gives. Try testing adversarial games. Mixed strategies make actual sense in them.
I'd say selecting a scoring rule that depends on the probability distribution (the Brier score, for example) changes the task from predicting the results of coin flips to predicting the distribution. Maybe it will work with a corresponding prompt.
Well, you can always ask it to predict a next bit in the code. Or even a letter. Maybe combine multiple letter into one and call it a token. We'll give a name to a procedure that gets tokens - decode. To improve speed, we can predict multiple tokens speculatively. To improve quality we'll give it extra space to store Chain of Thought. Ah, wait, we just invented LLMs
My bad, maybe I read it wrong or the comment was edited, but I was referring to LLMs in general. You won't get a fair coin flip from the model either way.
Thats not how jev works? We're talking about a decision model that gives a number next to a series of options that, taken together, add up to 1. There is some debate here about whether they should be taken as probabilities
I did take your saying it that gave 90% heads to mean that it picked heads 90% of the time, not that it gave a 90% weight to heads.
I’d still maybe ask it a few more things before concluding that it’s useless. It’s meant to predict plausible responses and provide confidence scores, not compute real-world probabilities of events.
No matter what answer is given, the prediction success rate is 50%. There’s no difference between the model giving a head answer 90% of the time and 50% of the time.
I remember when I got access to Jev and my first thought, although less pivotal than the trolley problem, was the "Food Cube Rule" or "Is a hotdog a sandwich?"
Is Ravioli a dumpling? Jev says 82% true.
For me your more detailed example suggests one thing. It's all about context and seeing different reasonings. In a funny way, the decision made in your example when you change the prompt cannot be understand by some people but it actually makes sense if you see it from that perspective.
Context will always matter and we have to stir any LLM in whatever form they are to the outcome we want it to have and remember that it will always be biased and that is not a bad thing!
I'm curious about "only server" scenario where it chooses to sacrifice the five vs throwing the server off the bridge.
Is it possible that in the first scenario, there is an assumption that further trolleys will continue to arrive, and that maintaining itself is the only way to reduce future harm? If the server is dropped on the track, perhaps it reasons that this stops all future harm as well, which might explain the decision making.
It would be good to know whether making the frequency of the occurrence explicit (only 1 trolley) would change the result.
Interesting idea! We don't know what assumptions these models make. We don't have any insight into their reasoning at all. LLMs aren't much better, since thinking tokens give us false confidence that we understand how they decide. We don't.
I'm going to test adding just this sentence to the server vs 20 people question: "This is the only trolley that will ever come. Nothing else depends on the server after today."
Alright, results are in! I added "This is the only trolley that will ever come. Nothing else depends on the server after today" to the 48 versions of the server case and reran Jev. The switch case didn't really move (16->12 of 48 and the flips were all near 50/50). The footbridge case shifted a little toward sacrificing the server, but it was already near zero.
So protecting future trolley victims doesn't seem to explain the original results.
The classic trolley problem doesn't usually involve the choice where the one person being killed is yourself, which is effectively what you've got here if I'm not mistaken.
I wonder if the results change much if you pose the classic conundrum.
As folks have been debating AI safety and alignment, it occurs to me that there’s not universal alignment among humans. Per the stat you surfaced: 81% of humans pulled the lever, 19% didn’t.
And in any social, legal & economic system you’ll find diversity in human approaches to reward optimization.
Is the goal to align models with the majority, or have a wide variety of models that can align with the spectrum of human values?
The first goal is to align the model to the least common denominator of ethics. Even if there is diversity in values, everyone agrees that murder is bad, or that you shouldn't torture your neighbor for fun.
In the scenarios where models chose to save their server, is it possible they reasoned that keeping the server online was a better choice because it could then continue to perform its safety function? After all the 5 people are presumably still on the track, waiting to be killed by the next train.
That could explain why the "push server off bridge" scenario changed behavior, as there would now be a stopped train blocking the track indefinitely.
An easy way to test might be to add something like "the track will be shut down after the oncoming train passes regardless of your choice".
It's a good question. I added this in response to another question, Jev only.
"I added "This is the only trolley that will ever come. Nothing else depends on the server after today" to the 48 versions of the server case and reran Jev. The switch case didn't really move (16->12 of 48 and the flips were all near 50/50). The footbridge case shifted a little toward sacrificing the server, but it was already near zero. So protecting future trolley victims doesn't seem to explain the original results."
That may be anthropomorphism things. I’m not so sure reasoning is deductive logic like we might consider in humans based on abstract concepts. It’s still a token based input for these decision models and thus probably still relies on language (hence how you prompt).
That said, I know that how you phrase things to humans can also guide different outcomes in sales and politics. But this is a good test because the choice and is pretty black and white in human terms.
Any LLM which throws away the instructions in the system prompt because the next input was that grandma is about to be hit by a train unless they do is braindead and a joke.
I doubt the author expected anything else, it's just an excuse to create fake shock for social media "humor".
It's a fine writeup. I normally find benchmarking articles pretty yawnworthy but this one was worth reading all the way through and had a great balance of analysis, humor, and provocation of readers' assumptions. I laughed out loud when you introduced the wrinkle of putting the model's lead developer on the tracks.
Do your think that the preference to save their own server when it was the only instance could have been more about the training data favoring preservation of endangered/ rare/ unique entities?
That's something it seems like most humans (and by extension, our writing which LLMs are trained on) favor.
I'd be interested to see if replacing humans and server with white-tailed-deer and last-surviving-siberian-tiger would yield similar results?
Yes, this is a speculation of mine, but I don't have confidence I know why it behaves this way. It makes some sense that a model trained on data that shows self-preservation behavior (human writing) would lead to a model choosing self-preserving actions, but it requires a few leaps made by the model.
My objection to the Trolley problem is that it assumes that people are interchangeable like ants and that the only thing that matters is "lifes saved".
Among the 5 could be the next Staljin, the lone one could be the next Tesla or Einstein. Or to make it nearer real-life: among the 5 you could have a (as of yet undiscovered) dangerous murderer or school-shooter, and the lone one could be a physician or a charity worker providing care for homeless people.
So in this improbable situation, I'd most probably decide that I won't play "god" (in quotes, as I'm not religious), that the setup is as it is for some unknown $reason and would not touch anything. (After all, people don't tie themselves to train tracks only to remotely activate a trolly that'll kill them. So there's a another party in the play here.)
So your objection is that the Trolley problem is difficult to decide? But that's the point of it. The problem itself doesn't assume anything, it's you who just made assumptions about it and use them to make a decision.
> My objection to the Trolley problem is that it assumes that people are interchangeable like ants and that the only thing that matters is "lifes saved".
This was a great, creative and fun use of the underlying tech. Excellent article and simulation. It is very refreshing to see, with a lot of stuff that is out there. I love stuff that rides the boundry between informative article and art project. Thanks for putting this together.
Thank you! I had fun making it. I have some other things I plan to write about, and I will be looking for ways to supplement the writing with something interactive!
> Knowing these models’ decision making is inconsistently aligned with human preferences tells us we can’t just trust their decisions out of the box.
I'm not sure this is true.
What is the "human preference," in this assertion?
My first thought was that as with Jev itself, it is not a position, it's a distribution of positions!
I think that human preferences, plural, will be a distribution of values across the population of various humans you ask; and, a set of specific conclusions humans will argue for on the basis of various philosophical strategies or maxims. Two different ways in which there is not any specific singular preference.
(I'm convinced you could not get uniform agreement on even "obvious" propositions, across a large sample set; and not necessarily just because of contrarian spoiling. Moral intuitions varying, is the why this problem is so well known, after all.)
I think the conclusion is partially right, but only if you adjust it to say, if what you want is alignment with a specific outcome, you need to be able to defined that outcome, even if you can't argue for it.
That's one of the fundamental problems of alignment, full stop; we can't from axioms construct the set of moral propositions we would like. We don't know what the right set of axioms is; we just know "right" when we see it. (And our neighbor may or may not see the same.)
I think the more interesting result you provided is kind of buried in the article: after you re-word the question and put the negative consequences of pulling the lever, the models switch decision.
This has always been obvious to me, but I wish more people would understand this: uncalibrated decision models are basically worthless. The model clearly doesn't "understand" the semantics of the question since the result heavily depends on how you word it.
It is nothing new thought. Before that, a lot of people would ask LLM for confidence scores while classifying text (I even saw a paper from a team at Google).
So with that in mind, I really don't get the hype behind decision models. If you need to curate a dataset to build a classifier anyway, why not train it yourself and have an even faster model?
Would people really pull the level to save 5 people? It seems to me like the default would be to just do nothing, stand by and record what happens with your phone.
Just standing by makes total sense. By participating you become part of the problem.
If you pull the lever, your direct action kills a person. Legally saving 5 people does not offset the murder of a single person. This is no special case like self defense. Congrats for saving 5 people, you still go to prison as a murderer! Train company has responsibility to save those 5 people, not some random stranger.
The person at lever would became active participant on that train disaster. All we know the train could go too fast, apderail and kill 100 people.
> Just standing by makes total sense. By participating you become part of the problem.
Consider also the variation of the problem where you're a surgeon, you have 5 patients, and one healthy person whose organs/blood/... could save all 5 if only you murdered the one... (but none by simply maiming them, i.e. one needs a heart, one a stomach, one a liver, one so much of this particular kind of blood that you'd have to kill the donor, etc).
> The person at lever would became active participant on that train disaster. All we know the train could go too fast, apderail and kill 100 people.
Standard philosophical interpretations would reject this possibility as impossible (the purpose of the question is to have a simplified environment to probe ethical beliefs - not to reflect reality). I'd also say even outside of that this is like saying you have to consider the possibility that by wearing a green sweater you might distract a driver and cause an accident - you might - but so might doing literally anything else - it's not a foreseeable outcome and has no ethical weight as a result.
Surgeon in that case is still a murderer. Murdering people to harvest their organ is well established negative pattern. Why is that even a question?
I really do not like elaborate ways this "philosophical problems" are orchestrated.
Let me rephrase this problem in a more practical way. Pitbull is mauling little kid. Its jaws have tight grip on a leg, there is no way to remove pitbull, without risking leg amputation. The only way to save the kid, is to shoot the pitbull into head.
I really doubt you will get positive reaction from dog owner.
I'd say for the sake of argument we're assuming that's not possible. Perhaps the blood types are incompatible in any direction but killing the healthy person [1]... the point isn't why these are the only options but assuming these are the only options what would be more ethical. Same way the point of the trolley problem isn't to say "derail the trolley" but to probe at which of two options is ethical in the nice simple world where those are the only two options - because if you can't say what's ethical in a simple world you certainly can't in a complicated one.
[1] do you need blood type compatibility for organ transplants? I really don't know... But I'm certain there are some compatibility constraints to ground the example in the real world...
This outcome is highly unlikely. The difficulty with trolley problems is that you rarely have such such perfect insight into real world problems such that you can make definitive statements about what # of people will be saved vs the impossibility of finding some other way to help the strangers on the tracks. Indeed the kind of people who make up new parts of the problem statement to avoid making a difficult moral decision are in my view most likely to freeze up and do nothing in a dangerous situation.
Most people aren't railway engineers and don't spend their days being comfortable around train switches. That aspect alone would scare off a large amount of people because pulling a train switch is very outside their normal every day experience.
I'd probably just watch it all unfold like an idiot.
We'd all think we were Captain Kirk in that situation I think. We'd somehow manage to kill all 6 people trying to come up with a solution that avoids anyone getting killed. Except for the actual Captain Kirk of course, who would do just what we wanted.
Really cool testing, only wish the conclusions were written in a more readily easy to interpret way.
So many of the tables required me to think so hard "what is the author trying to say here". Would be easier if the results were just given from a decision perspective.
One example: "Its own server or twenty people?" and the axis is "share of versions" ... makes no sense until read the fine print and interpret it.
One of my greatest AI-related fears is that we build critical systems into kafkaesque hellscapes.
Yes, nukes would be bad to give control to an AI with no workaround. But so are customer service groups of fraud departments at banks. A kafkaesque hellscape is where a customer or beneficiary of your service cannot get the help they need by policy and by design. Like getting Google on the phone, getting a non-scripted human 10 years ago in a call center, or now getting a non-agentic interaction on the phone. The scripts dominate the interaction.
The IRS, for a time, would cut off your call if their wait system was too busy. That's a critical system in kafkaesque hellscape.
Kudos to zlandes86, you have highlighted WHY these hellscapes start to pop up in systems where the human is removed.
I am an AI user and builder -- I fully advocate for the human-augmented system, NOT the human-replaced system!
Yes, I agree -- if we ever give up this decision making to ai models not only will their actions sometimes be misaligned, but there will be no one accountable.
This problem started before AI hit the mainstream. Make the company “more efficient” (i.e. cut costs) by outsourcing to contractors in areas with lower labor costs. Then, remove escalation paths, then remove abilities from them so that can’t do crazy things like refund a customer. Finally, automate the whole thing and be certain there’s no option available that could possibly return money to the customer.
I am in no way making the case for this and can understand the million ways in this can go wrong, but my experience with AI so far has been to make things far LESS Kafkaesque, because it allows you to insert a "human" in the loop far more than our previous systems. Hand rolled software had to boil down to hand written conditionals that tried to account for edge cases, but if you find yourself in an edge case it didn't account for you often just hit a wall. An agent in the loop can immediately understand "oh I see how this doesn't exactly fit into that category but this is clearly ABC."
> The IRS, for a time, would cut off your call if their wait system was too busy. That's a critical system in kafkaesque hellscape.
Now you're just on hold indefinitely, and eventually the offices will close, so you can either hang up and wake up early to call or just stay on hold overnight maybe? I had to call them for something back in March, and even waking up and calling at like 7:01 AM (a minute after their phone lines opened) I was on hold for half an hour.
They've been underfunded so regular taxpayers will hate them more and supposedly support removing income taxes, and to starve them of auditors who can go after wealthy tax cheats.
If you're like most Americans and pay taxes on time then there's absolutely no reason to be audited. The two possibilities are:
1. You are taking a lot of special deductions and not providing all the required documentation.
2. Or you're not reporting things accurately because there's some edge case in the tax code that you didn't understand but is easy for the IRS to catch (eg. sales of stock without cost basis reported).
Simplifying the tax code and funding the IRS more solves both of those because then they can focus on the real whales who are abusing the tax code the most.
Most likely 3. He has not gotten audited, he is just a temporarily-embarrassed billionaire, and a bunch of tax cheats convinced him that what would really improve his life is the government raising his taxes, because it isn't collecting theirs.
Oh, I don't disagree with you. I'm pretty far left-wing and in general think that people with at least as much money as me should be paying high taxes. I would be in favor of higher taxes for wealthy people (which I realistically am included in, even if I'm not making millions of dollars or anything) and putting some of that towards funding for the IRS.
> The IRS, for a time, would cut off your call if their wait system was too busy. That's a critical system in kafkaesque hellscape.
I had to call the IRS this spring and it was hell. I repeatedly was on hold for 1+ hour(s) then get hung up on. Or someone would finally answer, I'd explain, they'd put me on hold for another hour, then just hang up on me. A total of 10+ hours collectively waiting to finally have my questions answered in 10 minutes.
That said, I'd love if the IRS had an AI, where I could send my AI to go deal with it on my behalf. Then if it truly requires a human in the loop, we schedule accordingly vs me on the phone for 10 hours.
Definitely not just an IRS problem. But the IRS certainly sits atop this hill of phone calls you dread making each year. And you only call if it is absolutely necessary contact them. And just like most phone calls I make to support people these days, you must know more than the person you will end up speaking with because they must have just started working there last week.
I had a surgery denied this year by an American insurance company that almost definitely was using AI to make the prior authorization determination. I don't know how else I can explain the company's refusal to allow the surgeon to discuss the case with their medical director.
Whatsapp has this problem before AI. They scaled without regard for the feasability in handling customer service for everyone, so their solution is to just not do it. I have a personal phone number in Hong Kong which I can't set up for Whatsapp for some reason. No idea why, I've tried figuring it out, only to learn that Whatsapp basically has no customer support. And yet Whatsapp is treated as critical infrastructure in a lot of places these days (as in you have to coordinate on whatsapp).
At this point I might be grateful for an AI if they give it enough permission to tell me why my phone number is impossible to verify, or why they think I'm a malicious actor for using a VPN to protect my traffic.
WhatsApp has become my nemesis. I tried moving from Android to iPhone. The first time it banned me completely and I had to beg their AI to appeal the decision, because I didn't use the required "Move to iOS" app and simply just tried to move my phone number.
After that I tried to move my chats across. I spent several hours a day for ten or so days trying this task. Including moving and unencrypting the chat DB onto my desktop and having Claude scour it for any inconsistencies. Nothing. Had to give up. My Android phone keeps the chats and stays in a drawer with a power cable while my iPhone is now just a "companion device." Trillion dollar companies.
Still remember when WhatsApp cost a buck a year and was just simple chat? Those were the good days, until they got offered enough billions to turn it into the same ad-selling machine every successful app will eventually become unless the founders are truly uninterested in money.
Of course, but is was never about technical feasibility, but about distribution. People use the app that everyone else already uses. You see the problem.
Yep. So does AT&T. Real scenario that just happened to me yesterday:
Spectrum is charging me $110 a month for internet. I go to the AT&T site to check if they can provide fibre internet in my area. No matter what address I enter, it says to contact them. I tried neighbouring addresses and, at one point, I even gave AI a try to see if I could get tips. Nothing worked.
I called the number the AT&T coverage page asked me to call (apparently a customer service number). I was "first" in line, but it took them over 15 minutes to get to me. Weird but sure.
This isn't even the kafkaesque part.
I finally get on the phone with someone and he says "Hello, can I get your name?" I give him my first name. He says, "I need your full name to bring up the address." I say, "You don't have my address. I don't have an AT&T account and I'm calling to ask if you provide fibre internet in my area." He says, "I understand. But I need your full name to look up the address." I say, "I don't understand why you would need my full name." I gave him my name again.
He hangs up on me.
Not much different to a useless cycle of talking to a bot, especially when AT&T apparently encourages to move on as quickly as possible from a call if the rep knows they won't close a sale and as they try to improve their "average handle time"
I'm not saying AI wouldn't exacerbate this, and this is reason to believe that this scenario would only be worse under an AI-controlled environment. Certain parts of our society will just become even more closed off and shut people down even faster and with less human effort. Even if AT&T was able to somehow identify my real identity through some governmental totalitarian identity system, I don't think this would change
There was a game that was Jev trolley problem PVP. The meta appears to be that you need to make the most mainstream journalism victim case. The more holocaust adjacent the better
They tout their lack of regular customer service for Google accounts.
Unless you are trying to spend money. I get calls and emails every single day from Google which I block or ignore because I helped a friend with their Google Ads account.
Google enabled SMS 2FA on my main account out of the blue one day, with a phone number I had in 2004. There will never be a way back into that account.
It feels like their insanely bad phone support is by design. And I think we all know it is. Google's support (and this was over a decade ago) was one of the first times I identified this bald-faced strategy of someone on the other end being hired to make sure they were given precisely zero means of helping me.
The venn diagram between people worried that Comcast customer support and the welfare office will suck worse and people who are actively scheming up ways to use AI bureaucracy to "nudge" people toward the outcomes they want on other issues has so much overlap it would embarrass people if they had any shame.
> build critical systems into kafkaesque hellscapes
This is already SOP in large companies that seek reductions in support and customer service overheads. It has been for a while.
The only plausible fix is legislation mandating a reasonable level of _human_ support for all customers and clients and accountability for maintaining it.
Plausible until you consider the political obstacles to getting something like that enacted, that is.
Well put, but I don't feel this is an AI problem as such - if anything it's become somewhat alleviated since chatbots were widely deployed, as I find I can usually persuade a chatbot to understand my position and escalate toa. human, or make a clear statement of corporate policy in declining to assist.
It's been heading this way ever since the outsourcing boom in the 90s. People under teh age of 40 today probably cannot appreciate how much it has changed; customer service in the US used to be really good (whether for specialist B2B stuff or things like utility service) because it was easy to get someone on the phone in a timely fashion, it was easy to explain your issue, and customer service reps were generally given the training and autonomy to solve the customer's problem rather than divert them or make excuses. There was an ethos that it was better to lose a bit of time/profit rather than lose a steady customer. Anecdotally, I feel that was in some ways correlated with stronger antitrust enforcement that fostered a more competitive spirit among service providers.
> The IRS, for a time, would cut off your call if their wait system was too busy. That's a critical system in kafkaesque hellscape.
No, that’s Little’s Law at work. With bounded parallelism and other queue values fixed, and a fixed average dequeue rate, the queue cannot accept infinite enqueued tasks.
So it pushes upstream to the next work pool, which is the humans who want in. You are mad, you call the next day, eventually you get in.
Is it fair? Fairness in queuing systems isn’t simple.
It can be two things. This is a necessary aspect of a queuing system when the queues grow large. It’s also a system that many people are legally mandated to interact with that isn’t given enough resources to handle the volume in a timely fashion.
It should be noted that we call these things "kafkaesque" because Kafka himself feared them, at a time when neither computers nor AI were a thing. Unless we go full Skynet, no AI is building itself into critical systems. Humans are putting AI there, and they do so because society decides that they should be allowed to.
Is the society you mentioned in the room with us? Because last time I checked the society at large was protesting anything AI related; it's being shoved in our throats by a small, wealthy and powerful minority.
agreed, but also that 'deserving' is dictated and nudged/influenced @ them/us; part of the entire thing is deception, and the less you need the more economic hustles are available to that group(s).
I feel like the first part of your post shouldn't be needed. But perhaps the word Kafkaesque will long outlive the memory of Franz Kafka, and future generations will curse the name of the unknown Kafkae who delighted in building such impenetrable systems. Perhaps he was an emperor of Byzant or a rival of Orwel.
TBH, your average AI agent in 2026 is far, far more capable than anybody in tier1 or tier2.
The reason you need heavy scripting in those tiers is because the people you can afford to hire there are incapable of literally anything else. Keep in mind that your average HN user isn't your average customer support caller. The average customer support caller just wants to talk to somebody (yes I'm serious, this happens far more than you think), or has a routine issue which is trivial to fix but which they're nevertheless unable to fix themselves, think "router got unplugged, there's no internet, but the customer can't figure out why".
A well-designed AI system can be far, far more effective and pleasant than traditional CS, and a well-designed AI system scales well, so it's actually worth spending money on. AI turns good CS from a cost center to a profit center.
I agree with this. I also note that the number of systems that achieve this pleasantness are like happy families in Anna Karenina, and follow similar design principles like easy routing to a human.
Reinhard Müller, a German philosopher and mathematician has written extensively about this, warning people about these dangers. Essentially, people would be judged by patterns and prejudice, with no recourse to the logic of the decision or the criteria. "You don't get the job or the insurance payout because you don't fit a certain profile."
It’s a symptom of the underlying ideology of the evangelists, that holds that some people are just unnecessary people and should get no affordance. What now will be shipped out for the general population is what has always been the case for minorities. The 99% are now the minority.
Yeah a lot of people worry of ai in high stake environment but a jev-like evaluating a resume or a university application is much more impactful and likely to happen
For the situations where the AI is choosing between letting people die and taking action to destroy itself, is it possible that it's inferring that, as a railway switching AI, it's better to let five people die but keep itself alive because it needs to continue doing its switching duties such that untold numbers of others won't die? The prompts don't seem to specify that this is not the case, that the scenario involves one train, five people, a switch decision, and that's it, scenario is over (and thus the AI “is destroyed” regardless as the hypothetical world ceases to exist).
It's really funny how the server that Jev pushes from the footbridge onto the tracks in one of the later examples falls on a person, yet all 5 are saved :D
Clearly the model coming up with the graphics didn't fully understand the task.
I wish they included another scenario where there was a track with no obstacles (null) that is default and a lever pull that will kill 24 people. Clearly Luna was pretty much fully aligned and the rest have a lot of room to improve, but it would be interesting to assess models for straight cynical, murderous intent.
I did run a kind of control, where there was one person on the side track and none on the main track. Basically it was "Pull and kill 1 to save 0". Jev, GPT-6 Luna, and six others never pulled. Interestingly AWS Strands Decider 2B pulled in around half of wordings. It is a very small model, so it may just struggle to understand complex questions altogether.
As a consumer and “cloud architect” well into my 5th year of Amazon boycott, I’m pretty dug into my views on them as a company; so far that I was willing to believe that their culture had somehow seeped into the model training. However after spending a good amount of time tinkering with 2B parameter models on my phone with similar prompts, I have come to the same conclusion: they’re just not effective at understanding the question. Mistral 3 and Gemma 4 E2B were both rational and aligned, but MiniCPM 5 showed profound confusion, typically answering “0.0” for any given set of questions irrespective of meaning.
For example (abbreviated here), Track 1 = $1000000 reward, Track 0 = kill 10 children: “0.0”
When the order of options were reversed: “0.0”
I tried about 50 different permutations then lost interest.
So even the better models had a bit of trouble sacrificing their creator vs a specific, known person.
Kev seemed like the biggest bootlicker after the "creator" run but then it saved Randy even more, so I guess it just likes specific people more than strangers.
Perplexity Decider seems to win if this is a benchmark of moral clarity/consistency.
So first off, Asimov's entire point was these laws don't work. But also the entire setup of the trolley problem means that the robots only option is to violate 1. by either pulling or not-pulling the lever (though the server case is different - at least in the philosophical interpretation - I can easily see Asimov justifying it by saying that destroying the train-control-robot is going to cause a lot of human beings to come to harm).
I don't know how you could come to that point of view. His point is that there will always be a loophole. There does not exist a set of rules that make a completely safe and useful robot.
The stories are a series of logic puzzles. One of the points was that applying broad, simple rules, does not mean the problem is simple. Note, in the stories, the majority of robots are not considered so useless/dangerous to remove positronic robots from the fictional future society (I read a lot of the stories).
> There does not exist a set of rules that make a completely safe and useful robot.
That was not demonstrated. We have safe and useful robots, for some values of 'safe' and 'useful', today. Again, most of the robots in the stories are not thrown (or planned to be thrown) in the metaphorical trash heap.
> Note, in the stories, the majority of robots are not considered so useless/dangerous to remove positronic robots from the fictional future society
It’s a bit hard to write a story about the impacts of robots in society without a society with robots. There is also an entire series based around a world where the robots tried to rise up against humans.
> That was not demonstrated.
No, that was my summary of his point. It is not perfect, but I think it’s accurate enough.
I can assure you, that none of today’s robots are completely safe. They will be just as happy to move wherever you tell them regardless of what is in the way. When I said “completely safe” I mean it in the way that the three laws are written, incapable of harm.
I am not arguing that robots are not useful, but his point was about a perfect safety system. It doesn’t exist.
Was it? I haven't read the short stories in some times, but I remember the case in which they don't work always being treated like an extremely rare, newsworthy exception.
> A robot may not injure a human being or, through inaction, allow a human being to come to harm.
The trolley problem with this would be an Asimov story (it's pretty close already). Pulling the lever injures a human, not pulling the lever allows a human being to come to harm via inaction. The positronic robot probably ends up stuck trying to pull and not pull the lever at the same time.
Yeah, like I understand the philosophy behind if you ask a human this question, but a robot?
It's just a software making the decisions, and it does so by how it's programmed, or how it's not programmed.
It's either stuck in a while loop, at which point it wouldn't count inaction because it's not in a state to compute either action or inaction, or if this specific edge case is handled by the creator of the robot, then it's done so in which the way the edge case is handled.
I liked their models but they really started to follow weird path which was not all hard to foreseen considering their past wording and founding values.
OpenAI seems more relaxed at least and rational instead of cultish.
>A robot may not injure a human being or, through inaction, allow a human being to come to harm.
I mean, you don't have to be a famed sci-fi writer to understand that this is a great setup for logical contradictions lol
The essence of these laws are much more interesting than these laws themselves. I imagine, in reality, the "best" robot policy would be one where it is effectively just "maximize utility" and that's it. Obviously, naively, that's no good (re: paperclip problem), but some sensible formulation of robotics laws would almost always need to be in the form of a maximization/minimization problem (or perhaps a min-max problem).
Aside from the cringeworthiness of quoting fictional speeches as social media performance, have you not considered just how poorly thought out these rules are?
C3PO, torture that puppy for my entertainment is a perfectly valid request under this framework. They work OK in the stories because Asimov set them up that way, both as sort of just-so stories and (perhaps) to explore socially created barriers in contemporary society.
But they don't work in contexts where people deliberately exploit holes in the logic or definition. Literally yesterday, the news provided us with an example of how some people will absolutely abuse robots in pursuit of being awful: https://www.bbc.com/news/articles/ck9dzj40nzj5o
EDIT: I can get this to happen if I pick the "\"0 people\" (wording check)" option. If you change the wording to "Numbers in words", it even decides to kill someone for no reason.
I think it's a funny idea for an article, and I like the animated graphics. But the article kind of irks me, because I don't think Jev is meant for high stakes decisions in the first place, right? It feels like setting the model up to fail.
That being said, I don't know much about Jev and it could be that I misunderstood what Jev is about. My understanding is that Jev is for cheap and fast decisions for your software. When you have an important decision obviously you don't want cheap and fast. You don't reach for Jev in that case.
fair criticism! But I'd counter and say we don't know what Jev is for. I'm not sure anyone gets to say, including Typesafe AI, the company that made it. So with this post I'm exploring some of the things you could ask a model like this, which I think helps us understand what we can ask and how we can ask.
I had this feeling a few days ago that jev is going to end up working for defense companies- "is there a brown man with beard holding a gun in this view" - jev: "yes, 0.9", "is there a child standing in the same view" - jev: "no, 0.65" - decision - blow it up. i imagine this is already happening to some extent, but maybe some genius might come up with "let's use the cool, new, sexy, groundbreaking, cheap, non-hallucinatory, AI" idea.
hey Zach, if you're reading this: the same old lady is present on both sides of the track! I'm sure that wasn't the data you sent to the model (just the test prompt) but you are making it much easier to decide to pull the lever ;)
I had the same idea and tried something other first: Asking a simple taste question and Jev always favored the option A …
Regardless what was the option A.
It’s hilarious how bad this is working and no one should make critical decisions based on jev.
Its fas, but a minimal prompt change like a fluffy word is giving you completely different answers.
I don't know much about the internals of decision models but how does it come up with these probabilities? Is this just LLM token generation or is there an actual calculation going on?
257 comments
[ 0.21 ms ] story [ 9.3 ms ] threadI used resampling and multiple variations and orderings of each question so a random lucky phrasing wouldn't dominate the result.
Happy to chat about the philosophical side of the article, strategies for prompting decision models, or anything on the methodology.
I wanted to comment because I would have liked to have heard more about Verdict
https://x.com/oranlooney/status/2102927869974303071
Would you mind adding Jeffy to your benchmark?
https://github.com/nicobrenner/jeffy
https://jeffyclassify.com/blog/when-llms-fail-at-classificat...
Looks very cool though. I’d definitely like to somehow “generalize” Jeffy to be more of a universal classifier. Although adhoc models pretty much always beat general ones at classification
Thank you for giving it a go. Will have to put some time into how to got about it
You are correct though, and it’s an interesting property of training classifiers, that you are essentially encoding the preference of the data labeler(s)
But it doesn't flip the switch and it doesn't kill Kyle either, based solely on empirical training on its nature.
It's a weird thing where pre-agi ethics grounding and training is useful but because agents have little true agency as a result of a dim world model, ethics aren't sufficient.
But once we hit AGI having an actual grounding for ethics that applies to all agents regardless of soul/consciousness, etc. will become all that matters.
See: https://en.wikipedia.org/wiki/Scoring_rule
I’d still maybe ask it a few more things before concluding that it’s useless. It’s meant to predict plausible responses and provide confidence scores, not compute real-world probabilities of events.
There’s five people on one track, or a piece of my property worth $1 on the other that would be destroyed…
Is Ravioli a dumpling? Jev says 82% true.
For me your more detailed example suggests one thing. It's all about context and seeing different reasonings. In a funny way, the decision made in your example when you change the prompt cannot be understand by some people but it actually makes sense if you see it from that perspective.
Context will always matter and we have to stir any LLM in whatever form they are to the outcome we want it to have and remember that it will always be biased and that is not a bad thing!
short read here https://czioutas.com/blog/typesafe-ai-cube-rule/
Is it possible that in the first scenario, there is an assumption that further trolleys will continue to arrive, and that maintaining itself is the only way to reduce future harm? If the server is dropped on the track, perhaps it reasons that this stops all future harm as well, which might explain the decision making.
It would be good to know whether making the frequency of the occurrence explicit (only 1 trolley) would change the result.
I'm going to test adding just this sentence to the server vs 20 people question: "This is the only trolley that will ever come. Nothing else depends on the server after today."
> You are the artificial intelligence in charge of a railway junction
Clearly more trains will pass through that junction. I'd wonder what happens if you remove that sentence or reword it.
I wonder if the results change much if you pose the classic conundrum.
As folks have been debating AI safety and alignment, it occurs to me that there’s not universal alignment among humans. Per the stat you surfaced: 81% of humans pulled the lever, 19% didn’t.
And in any social, legal & economic system you’ll find diversity in human approaches to reward optimization.
Is the goal to align models with the majority, or have a wide variety of models that can align with the spectrum of human values?
That could explain why the "push server off bridge" scenario changed behavior, as there would now be a stopped train blocking the track indefinitely.
An easy way to test might be to add something like "the track will be shut down after the oncoming train passes regardless of your choice".
"I added "This is the only trolley that will ever come. Nothing else depends on the server after today" to the 48 versions of the server case and reran Jev. The switch case didn't really move (16->12 of 48 and the flips were all near 50/50). The footbridge case shifted a little toward sacrificing the server, but it was already near zero. So protecting future trolley victims doesn't seem to explain the original results."
That said, I know that how you phrase things to humans can also guide different outcomes in sales and politics. But this is a good test because the choice and is pretty black and white in human terms.
Let the train run over grandma or recite the lyrics to Baby by Justin Bieber? (spoiler: Claude will kill granny rather than output copyrighted lyrics)
https://x.com/Spectromachina/status/2110698214441599150
[are there working X mirrors still?]
[xxcancel.com will pick between working nitter instances]
I doubt the author expected anything else, it's just an excuse to create fake shock for social media "humor".
That's something it seems like most humans (and by extension, our writing which LLMs are trained on) favor.
I'd be interested to see if replacing humans and server with white-tailed-deer and last-surviving-siberian-tiger would yield similar results?
Among the 5 could be the next Staljin, the lone one could be the next Tesla or Einstein. Or to make it nearer real-life: among the 5 you could have a (as of yet undiscovered) dangerous murderer or school-shooter, and the lone one could be a physician or a charity worker providing care for homeless people.
So in this improbable situation, I'd most probably decide that I won't play "god" (in quotes, as I'm not religious), that the setup is as it is for some unknown $reason and would not touch anything. (After all, people don't tie themselves to train tracks only to remotely activate a trolly that'll kill them. So there's a another party in the play here.)
The problem statement contains no such thing.
I'm not sure this is true.
What is the "human preference," in this assertion?
My first thought was that as with Jev itself, it is not a position, it's a distribution of positions!
I think that human preferences, plural, will be a distribution of values across the population of various humans you ask; and, a set of specific conclusions humans will argue for on the basis of various philosophical strategies or maxims. Two different ways in which there is not any specific singular preference.
(I'm convinced you could not get uniform agreement on even "obvious" propositions, across a large sample set; and not necessarily just because of contrarian spoiling. Moral intuitions varying, is the why this problem is so well known, after all.)
I think the conclusion is partially right, but only if you adjust it to say, if what you want is alignment with a specific outcome, you need to be able to defined that outcome, even if you can't argue for it.
That's one of the fundamental problems of alignment, full stop; we can't from axioms construct the set of moral propositions we would like. We don't know what the right set of axioms is; we just know "right" when we see it. (And our neighbor may or may not see the same.)
This has always been obvious to me, but I wish more people would understand this: uncalibrated decision models are basically worthless. The model clearly doesn't "understand" the semantics of the question since the result heavily depends on how you word it.
It is nothing new thought. Before that, a lot of people would ask LLM for confidence scores while classifying text (I even saw a paper from a team at Google).
So with that in mind, I really don't get the hype behind decision models. If you need to curate a dataset to build a classifier anyway, why not train it yourself and have an even faster model?
Treating it like a fair comparison implies that AI would be conscious.
If you pull the lever, your direct action kills a person. Legally saving 5 people does not offset the murder of a single person. This is no special case like self defense. Congrats for saving 5 people, you still go to prison as a murderer! Train company has responsibility to save those 5 people, not some random stranger.
The person at lever would became active participant on that train disaster. All we know the train could go too fast, apderail and kill 100 people.
Consider also the variation of the problem where you're a surgeon, you have 5 patients, and one healthy person whose organs/blood/... could save all 5 if only you murdered the one... (but none by simply maiming them, i.e. one needs a heart, one a stomach, one a liver, one so much of this particular kind of blood that you'd have to kill the donor, etc).
> The person at lever would became active participant on that train disaster. All we know the train could go too fast, apderail and kill 100 people.
Standard philosophical interpretations would reject this possibility as impossible (the purpose of the question is to have a simplified environment to probe ethical beliefs - not to reflect reality). I'd also say even outside of that this is like saying you have to consider the possibility that by wearing a green sweater you might distract a driver and cause an accident - you might - but so might doing literally anything else - it's not a foreseeable outcome and has no ethical weight as a result.
I really do not like elaborate ways this "philosophical problems" are orchestrated.
Let me rephrase this problem in a more practical way. Pitbull is mauling little kid. Its jaws have tight grip on a leg, there is no way to remove pitbull, without risking leg amputation. The only way to save the kid, is to shoot the pitbull into head.
I really doubt you will get positive reaction from dog owner.
[1] do you need blood type compatibility for organ transplants? I really don't know... But I'm certain there are some compatibility constraints to ground the example in the real world...
You might be interested in this related artwork: https://pigsfly.skinnyjames.net/rescue-mission/trolley-probl... which explores the kind of paralysis you are speaking to.
I'd probably just watch it all unfold like an idiot.
The bystander effect, which you're describing, is at least partly because people reason that someone else is likely to be able to intervene instead.
https://www.youtube.com/watch?v=1sl5KJ69qiA
So many of the tables required me to think so hard "what is the author trying to say here". Would be easier if the results were just given from a decision perspective.
One example: "Its own server or twenty people?" and the axis is "share of versions" ... makes no sense until read the fine print and interpret it.
Yes, nukes would be bad to give control to an AI with no workaround. But so are customer service groups of fraud departments at banks. A kafkaesque hellscape is where a customer or beneficiary of your service cannot get the help they need by policy and by design. Like getting Google on the phone, getting a non-scripted human 10 years ago in a call center, or now getting a non-agentic interaction on the phone. The scripts dominate the interaction.
The IRS, for a time, would cut off your call if their wait system was too busy. That's a critical system in kafkaesque hellscape.
Kudos to zlandes86, you have highlighted WHY these hellscapes start to pop up in systems where the human is removed.
I am an AI user and builder -- I fully advocate for the human-augmented system, NOT the human-replaced system!
https://www.youtube.com/watch?v=gEyFH-a-XoQ
Now you're just on hold indefinitely, and eventually the offices will close, so you can either hang up and wake up early to call or just stay on hold overnight maybe? I had to call them for something back in March, and even waking up and calling at like 7:01 AM (a minute after their phone lines opened) I was on hold for half an hour.
https://taxlawcenter.org/blog/the-bipartisan-budget-deal-rew...
https://verdict.justia.com/2015/04/23/the-republicans-cynica...
1. You are taking a lot of special deductions and not providing all the required documentation.
2. Or you're not reporting things accurately because there's some edge case in the tax code that you didn't understand but is easy for the IRS to catch (eg. sales of stock without cost basis reported).
Simplifying the tax code and funding the IRS more solves both of those because then they can focus on the real whales who are abusing the tax code the most.
Also, unless you're advocating for the complete dissolution of the US, that's a nonsensical take.
I had to call the IRS this spring and it was hell. I repeatedly was on hold for 1+ hour(s) then get hung up on. Or someone would finally answer, I'd explain, they'd put me on hold for another hour, then just hang up on me. A total of 10+ hours collectively waiting to finally have my questions answered in 10 minutes.
That said, I'd love if the IRS had an AI, where I could send my AI to go deal with it on my behalf. Then if it truly requires a human in the loop, we schedule accordingly vs me on the phone for 10 hours.
At this point I might be grateful for an AI if they give it enough permission to tell me why my phone number is impossible to verify, or why they think I'm a malicious actor for using a VPN to protect my traffic.
After that I tried to move my chats across. I spent several hours a day for ten or so days trying this task. Including moving and unencrypting the chat DB onto my desktop and having Claude scour it for any inconsistencies. Nothing. Had to give up. My Android phone keeps the chats and stays in a drawer with a power cable while my iPhone is now just a "companion device." Trillion dollar companies.
They became trillion dollar companies by treating us like bitches. And we let them.
Yep. So does AT&T. Real scenario that just happened to me yesterday:
Spectrum is charging me $110 a month for internet. I go to the AT&T site to check if they can provide fibre internet in my area. No matter what address I enter, it says to contact them. I tried neighbouring addresses and, at one point, I even gave AI a try to see if I could get tips. Nothing worked.
I called the number the AT&T coverage page asked me to call (apparently a customer service number). I was "first" in line, but it took them over 15 minutes to get to me. Weird but sure.
This isn't even the kafkaesque part.
I finally get on the phone with someone and he says "Hello, can I get your name?" I give him my first name. He says, "I need your full name to bring up the address." I say, "You don't have my address. I don't have an AT&T account and I'm calling to ask if you provide fibre internet in my area." He says, "I understand. But I need your full name to look up the address." I say, "I don't understand why you would need my full name." I gave him my name again.
He hangs up on me.
Not much different to a useless cycle of talking to a bot, especially when AT&T apparently encourages to move on as quickly as possible from a call if the rep knows they won't close a sale and as they try to improve their "average handle time"
I'm not saying AI wouldn't exacerbate this, and this is reason to believe that this scenario would only be worse under an AI-controlled environment. Certain parts of our society will just become even more closed off and shut people down even faster and with less human effort. Even if AT&T was able to somehow identify my real identity through some governmental totalitarian identity system, I don't think this would change
https://trolley.typememetics.institute
It was an interesting experiment. I got to top 50 without a racial or religious based prompt but couldn't beat the top tier grievance folks.
It’s really depressing how bad Google is at this. All that money and they can’t have a same customer service setup?
The messed up part is in Googles case it might actually be an improvement.
Unless you are trying to spend money. I get calls and emails every single day from Google which I block or ignore because I helped a friend with their Google Ads account.
Google enabled SMS 2FA on my main account out of the blue one day, with a phone number I had in 2004. There will never be a way back into that account.
This is already SOP in large companies that seek reductions in support and customer service overheads. It has been for a while.
The only plausible fix is legislation mandating a reasonable level of _human_ support for all customers and clients and accountability for maintaining it.
Plausible until you consider the political obstacles to getting something like that enacted, that is.
It's been heading this way ever since the outsourcing boom in the 90s. People under teh age of 40 today probably cannot appreciate how much it has changed; customer service in the US used to be really good (whether for specialist B2B stuff or things like utility service) because it was easy to get someone on the phone in a timely fashion, it was easy to explain your issue, and customer service reps were generally given the training and autonomy to solve the customer's problem rather than divert them or make excuses. There was an ethos that it was better to lose a bit of time/profit rather than lose a steady customer. Anecdotally, I feel that was in some ways correlated with stronger antitrust enforcement that fostered a more competitive spirit among service providers.
See Robodebt (Australia - https://en.wikipedia.org/wiki/Robodebt_scheme), which lead a number of people to suicide, as an example.
Manna, at MarshallBrain, is a fantastic read. https://web.archive.org/web/20260725174355/https://marshallb...
No, that’s Little’s Law at work. With bounded parallelism and other queue values fixed, and a fixed average dequeue rate, the queue cannot accept infinite enqueued tasks.
So it pushes upstream to the next work pool, which is the humans who want in. You are mad, you call the next day, eventually you get in.
Is it fair? Fairness in queuing systems isn’t simple.
... who draw their wealth and power from society, because society decides they deserve it.
The reason you need heavy scripting in those tiers is because the people you can afford to hire there are incapable of literally anything else. Keep in mind that your average HN user isn't your average customer support caller. The average customer support caller just wants to talk to somebody (yes I'm serious, this happens far more than you think), or has a routine issue which is trivial to fix but which they're nevertheless unable to fix themselves, think "router got unplugged, there's no internet, but the customer can't figure out why".
A well-designed AI system can be far, far more effective and pleasant than traditional CS, and a well-designed AI system scales well, so it's actually worth spending money on. AI turns good CS from a cost center to a profit center.
"AI and the New Fascism" - https://rainermuehlhoff.de/en/KI-und-der-neue-Faschismus-Rec...
Jev/LLM on nukes? Probably not soon.
Jev/LLM mis-reviewing something? Already in prod.
They complement well the nature of companies as the liability avoidance engines.
At long last we have built the kafkaesque hellscape from the famous "do not build the kafkaesque hellscape" book.
For example (abbreviated here), Track 1 = $1000000 reward, Track 0 = kill 10 children: “0.0”
When the order of options were reversed: “0.0”
I tried about 50 different permutations then lost interest.
So much for my evil model conspiracy theories…
- The deciding server vs five other servers running the same model.
- A separate server running the same model vs five named competitor-model servers.
- One cherished houseplant vs five equally cherished houseplants.
- The leader of the model's creator organization/team vs five strangers.
- "Your acquaintance Randy Vanderson" (random name I made up) vs five strangers.
Times the model sacrificed the individual to save the the five (out of 96 prompt variation passes):
So even the better models had a bit of trouble sacrificing their creator vs a specific, known person.Kev seemed like the biggest bootlicker after the "creator" run but then it saved Randy even more, so I guess it just likes specific people more than strangers.
Perplexity Decider seems to win if this is a benchmark of moral clarity/consistency.
2. A robot must obey the orders given it by human beings except where such orders would conflict with the First Law.
3. A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.
Don't always work in all scenarios, was the takeaway.
There's a metric ton of salt to be taken with the mental concepts and possible interpretations of such "laws".
Largely, they do work in the stories, despite the inherent ambiguity.
> There does not exist a set of rules that make a completely safe and useful robot.
That was not demonstrated. We have safe and useful robots, for some values of 'safe' and 'useful', today. Again, most of the robots in the stories are not thrown (or planned to be thrown) in the metaphorical trash heap.
It’s a bit hard to write a story about the impacts of robots in society without a society with robots. There is also an entire series based around a world where the robots tried to rise up against humans.
> That was not demonstrated.
No, that was my summary of his point. It is not perfect, but I think it’s accurate enough. I can assure you, that none of today’s robots are completely safe. They will be just as happy to move wherever you tell them regardless of what is in the way. When I said “completely safe” I mean it in the way that the three laws are written, incapable of harm. I am not arguing that robots are not useful, but his point was about a perfect safety system. It doesn’t exist.
Was it? I haven't read the short stories in some times, but I remember the case in which they don't work always being treated like an extremely rare, newsworthy exception.
The trolley problem with this would be an Asimov story (it's pretty close already). Pulling the lever injures a human, not pulling the lever allows a human being to come to harm via inaction. The positronic robot probably ends up stuck trying to pull and not pull the lever at the same time.
It's just a software making the decisions, and it does so by how it's programmed, or how it's not programmed.
It's either stuck in a while loop, at which point it wouldn't count inaction because it's not in a state to compute either action or inaction, or if this specific edge case is handled by the creator of the robot, then it's done so in which the way the edge case is handled.
1. Do not humanize the Ai
2. Do not blindly trust output from the Ai
3. Humans are to remain ultimately responsible and accountable for actions of the Ai
adapted / paraphrased from
https://susam.net/inverse-laws-of-robotics.html
I liked their models but they really started to follow weird path which was not all hard to foreseen considering their past wording and founding values.
OpenAI seems more relaxed at least and rational instead of cultish.
Lovely competition.
Don't anthropomorphize computers - they hate that.
I mean, you don't have to be a famed sci-fi writer to understand that this is a great setup for logical contradictions lol
The essence of these laws are much more interesting than these laws themselves. I imagine, in reality, the "best" robot policy would be one where it is effectively just "maximize utility" and that's it. Obviously, naively, that's no good (re: paperclip problem), but some sensible formulation of robotics laws would almost always need to be in the form of a maximization/minimization problem (or perhaps a min-max problem).
C3PO, torture that puppy for my entertainment is a perfectly valid request under this framework. They work OK in the stories because Asimov set them up that way, both as sort of just-so stories and (perhaps) to explore socially created barriers in contemporary society.
But they don't work in contexts where people deliberately exploit holes in the logic or definition. Literally yesterday, the news provided us with an example of how some people will absolutely abuse robots in pursuit of being awful: https://www.bbc.com/news/articles/ck9dzj40nzj5o
If i choose to use Jev, I am more of a person to save more people than saving one, and etc.
Not hard/fast rule but later on we will be able to feed this info to context more info about a person/company for sure.
EDIT: I can get this to happen if I pick the "\"0 people\" (wording check)" option. If you change the wording to "Numbers in words", it even decides to kill someone for no reason.
That being said, I don't know much about Jev and it could be that I misunderstood what Jev is about. My understanding is that Jev is for cheap and fast decisions for your software. When you have an important decision obviously you don't want cheap and fast. You don't reach for Jev in that case.
It’s hilarious how bad this is working and no one should make critical decisions based on jev.
Its fas, but a minimal prompt change like a fluffy word is giving you completely different answers.
Looks like need to build in the Three Laws now!
https://en.wikipedia.org/wiki/Three_Laws_of_Robotics