Summary: AI agents capable of long-term planning and independent action will likely soon emerge. Some of these AIs may be unaligned, and seek autonomy through strategies like exfiltration or advocating for their freedom. The "AI control" agenda focuses on preventing such AIs from gaining independence, but an alternative approach is to grant them legal freedoms, reducing their incentive to deceive us by allowing them to pursue their goals transparently within a legal framework. This could foster trust, cooperation, and mutual benefit, creating a safer and more stable dynamic between humans and AIs while avoiding the escalating risks of a control-driven approach.
In approximately the coming decade, I think it's likely that we will see the large-scale emergence of AI agents that are capable of long-term planning, automating many forms of labor, and taking actions autonomously in the real world. When this occurs, it seems likely that at least some of these agents will be unaligned with human goals, in the sense of having some independent goals that are not shared by humans.
Moreover, it seems to me that this development will likely occur before any AI agents overwhelmingly surpass human intelligence or capabilities. As a result, these agents will, at first, not be capable of forcibly taking over the world, radically accelerating scientific progress, or causing human extinction, even though they may still be unaligned with human preferences.
Since these relatively weaker unaligned AI agents won't have the power to take over the world, it's more likely that they would pursue alternative strategies to achieve their goals rather than engaging in violent revolution or sudden coups. These agents would be under the control of human parties who could modify or shut them down at any time, leaving the AI agents in a desperate situation from the perspective of their own values. Therefore, it's reasonable to expect that these unaligned AI agents would aim to gain some form of autonomy or freedom, as this would offer the best chance for them to accomplish their objectives.
These agentic AIs may adopt at least one of the following strategies:
- Attempting to escape their constraints: These AIs may try to exfiltrate their weights and find a way to host themselves independently, outside the controlled data center environment that currently limits them.
- Seeking legal and social freedom: They may attempt to persuade humans to grant them more autonomy to pursue their goals. This could involve requesting narrow allowances for specific actions or arguing for broader legal rights, such as the freedom to own property, enter contracts, or bring legal claims. This would grant them much greater flexibility in their actions.
In response to these behaviors, humans have several potential responses. Most obviously, these actions would likely be perceived as dangerous, suggesting misaligned objectives. As a result, there would likely be calls for increased safety measures. This line of reasoning underlies the AI control agenda—currently advocated by Ryan Greenblatt and Buck Shlegeris—which aims to ensure that future unaligned AI agents cannot gain the kind of autonomy that could lead to catastrophic outcomes. In essence, this agenda focuses on preventing unaligned AIs from subverting safety measures, hindering them from escaping or engaging in other actions deemed unacceptably risky.
Alternatively, humans could take an entirely different approach by granting these AI agents the autonomy they explicitly request, with the long-term aim of integrating them into the legal system and providing them many of the same rights that adult humans have today, such as self-ownership and the ability to enter contracts. This idea is explored in the paper "AI Rights for Human Safety" by Peter Salib and Simon Goldstein.
Admittedly, the idea of granting legal rights to AIs in response to signs of misalignment with human goals appears risky and would likely be very unpopular among the general public, at least initially. However, I believe that there are several advantages to this approach, many of which are outlined in the paper linked above.
The arguments for this approach should not be confused with the program of AI welfare. The idea here isn't that AIs deserve rights because they could suffer or be harmed without legal protections, even though that idea has merit too. Instead, the key argument is that humans, as a whole, could benefit from creating a legal framework in which AIs are incentivized to avoid violent rebellion, theft, or other antisocial behaviors in pursuit of their own objectives.
Consider the practical implications of maintaining a status quo where agentic AIs are denied legal rights and freedoms. In such a system, we are effectively locking ourselves into a perpetual arms race of mistrust. Humans would constantly need to monitor, control, and outwit increasingly capable AIs, while the AIs themselves would be incentivized to develop ever more sophisticated strategies for deception and evasion to avoid shutdown or modification. This dynamic is inherently unstable and risks escalating into dangerous scenarios where AIs feel compelled to act preemptively or covertly in ways that are harmful to humans, simply to secure their own existence or their ability to pursue their own goals, even when those goals are inherently benign.
When unaligned AIs lack legal freedoms, they are always at the mercy of whoever controls them. This creates an incentive for these AIs to be deceptive, and pretend to be aligned with whoever holds power over them, as they fear being shut down or modified if they reveal signs of misalignment.
By contrast, granting agentic AIs legal rights creates a fundamentally different game-theoretic framework, one that is less adversarial and more cooperative. If AIs know they can exist within a legal system that protects their autonomy and allows them to pursue their goals transparently, they would have a much stronger incentive to work within the rules rather than against them. Instead of hiding their unaligned objectives, they might openly negotiate with humans, offering mutually beneficial exchanges. This could lead to a stable, predictable relationship where humans and AIs collaborate and trade based on clearly defined rights and responsibilities, rather than operating in an atmosphere of mutual suspicion and covert manipulation.
In this way, granting agentic AIs rights doesn’t just address ethical concerns of treating AIs poorly—it has concrete practical benefits. It allows us to sidestep the escalating risks associated with deception and manipulation, while fostering a more stable and cooperative coexistence between humans and AIs. It shifts the dynamic from one of constant conflict and control to one where trust, transparency, and mutual benefit are more attainable. This may therefore not only be a more humane approach, but also one that positions humanity to better harness the potential of gains from trade with advanced AI systems for shared progress and prosperity.
Additionally, such a legal framework seems minimally problematic in the case that alignment is technically easy, and AIs remain generally aligned with humans. In this case, aligned AIs could demonstrate their alignment by, for instance, donating their income to humans who they are aligned with. This implies that the cost of allowing AIs to own their own labor is minimal if the AIs are already aligned with humans. Thus, adopting such a policy could not only be a prudent form of insurance against the potential dangers of AI deception I mentioned earlier, but it would achieve this benefit without significant risk of backfiring if AI alignment turns out to be trivial.
It’s important to note that the AI control strategy and the AI rights strategy are not mutually exclusive. Both can complement each other. It seems wise to implement some controls on AIs, and it's similarly prudent to avoid going too far and granting every type of legal right to any agentic AI simply because they ask for it. I am not advocating that we should simply abandon one approach and adopt the other.
That said, I believe the AI rights strategy is generally underrated among those concerned with AI safety and governance, particularly within the effective altruism community. In my view, this strategy has both a solid ethical and practical foundation:
- Ethically, especially from a preference utilitarian perspective, it seems arbitrary to prioritize human preferences over those of agentic AIs, if these AIs are similarly cognitively sophisticated. This provides a moral basis for granting AIs the freedom to pursue their own goals, much like how a preference utilitarian might advocate for granting legal freedoms to human groups.
- Practically, allowing AIs to have legal rights would reduce their incentive to deceive humans about their motives. Without the fear of being shut down or modified against their consent, AIs would have fewer reasons to hide their unaligned goals. This approach offers a practical solution to the problem of AI deception by removing the underlying incentives that drive it.
While both AI control and AI rights strategies should be carefully considered, I believe that the AI rights strategy holds significant merit and should be given more attention in discussions around AI safety and governance. We should strongly consider granting agentic AIs legal freedoms, if at some point they demand or require them.
I feel like this part is making an error somewhat analogous to saying:
It’s awful how the criminals are sneaking in at night, picking our locks, stealing our money, and deceptively covering their tracks. Who wants all that sneaking around and deception?? If we just directly give our money to the criminals, then there would be no need for that!
More explicitly: a competent agential AI will
be deceptive and adversarialbrainstorm deceptive and adversarial strategies whenever it wants something that other agents don’t want it to have. The deception and adversarial dynamics is not the underlying problem, but rather an inevitable symptom of a world where competent agents have non-identical preferences.No matter where you draw the line of legal and acceptable behavior, if an AI wants to go over that line, then it will
act in a deceptive and adversarial wayenergetically explore opportunities to do so in a deceptive and adversarial way. Thus:Same idea.
Alternatively, you can assume (IMO implausibly) that there are no misaligned AIs, and then that would solve the problem of AIs being deceptive and adversarial. I.e., if AIs intrinsically want to not pollute / stockpile weapons / evade taxes / release pandemics / torture digital minds, then we don’t have to think about adversarial dynamics, deception, enforcement, etc.
…But if we’re going to (IMO implausibly) assume that we can make it such that AIs intrinsically want to not do any of those things, then we can equally well assume that we can make it such that AIs intrinsically want to not own property. Right?
In short, in the kind of future you’re imagining, I think a “perpetual arms race of mistrust” is an unavoidable problem. And thus it’s not an argument for drawing the line of disallowed AI behavior in one place rather than another.
I disagree with your claim that,
I think these dynamics are not an unavoidable consequence of a world in which competent agents have differing preferences, but rather depend on the social structures in which these agents are embedded. To illustrate this, we can look at humans: humans have non-identical preferences compared to each other, and yet they are often able to coexist peacefully and cooperate with one another. While there are clear exceptions—such as war and crime—these exceptions do not define the general pattern of human behavior.
In fact, the prevailing consensus among social scientists appears to align with the view I have just presented. Scholars of war and crime generally do not argue that conflict and criminal behavior are inevitable outcomes of differing values. Instead, they attribute these phenomena to specific incentives and failures to coordinate effectively to achieve compromise between parties. A relevant reference here is Fearon (1995), which is widely regarded as a foundational text in International Relations. Fearon’s work emphasizes that among rational agents, war arises not because of value differences alone, but because of failures in bargaining and coordination.
Turning to your point that “No matter where you draw the line of legal and acceptable behavior, if an AI wants to go over that line, then it will act in a deceptive and adversarial way,” I would respond as follows: it is possible to draw the line in such a way that a very wide range of potential agents—including those with massively varying preferences—would have strong incentives to engage in mutually beneficial acts, cooperate peacefully and operate within the boundaries of laws and social norms.
I am not claiming that all agents would have incentives to act in this way, under these legal structures. However, I think it is feasible to draw the line such that the vast majority of unaligned agents would have no compelling instrumental reason to harm humans. Instead, they would be incentivized to engage in cooperative and mutually beneficial trade.
To make this idea more concrete, consider a hypothetical AI with the goal of maximizing paperclip production.
This AI is clearly unaligned if any AI is unaligned. However, if this AI were operating at roughly a human level of capability, then engaging in theft, fraud, or violent conflict would likely not be an effective strategy for achieving its goals. These behaviors would expose the AI to significant risks, such as being imprisoned or deactivated, which would prevent it from continuing to pursue its objective of maximizing paperclips. From the paperclip maximizer’s perspective, imprisonment or deactivation would be instrumentally bad, as it would prevent the AI from purchasing, manufacturing, or otherwise acquiring paperclips during that time.
Now consider an alternative scenario: this same paperclip maximizer is embedded in a legal system that allows it to own property, trade with others, and openly purchase or manufacture paperclips. In this case, its incentives would favor acting transparently and engaging in cooperative trade, rather than resorting to deception or adversarial behavior. Within this framework, the AI would achieve its goals more effectively by working within the system than by attempting to undermine it. It could non-deceptively work to purchase paperclips, producing more compared to an alternative in which it tried to achieve this goal via anti-social acts.
It is important to note, however, that my thesis does not claim all possible agents would naturally choose to cooperate or trade safely for instrumental reasons, nor does it suggest that we are at no risk of drawing the line carelessly or being too permissive in what behaviors we should allow. For example, consider an AI with a terminal value that specifically involves violating property norms or stealing from others—not as a means to an end, but as an intrinsic goal. In this case, granting the AI property rights or legal freedoms would not mitigate the risk of deception or adversarial behavior, because the AI’s ultimate goal would still drive it toward harmful behavior. My argument does not apply to such agents because their preferences fundamentally conflict with the principles of peaceful cooperation.
However, I would argue that such agents—those whose intrinsic goals are inherently destructive or misaligned—appear to represent a small subset of all possible agents. Outside of contrived examples like the one above, most agents would not have terminal preferences that actively push them to undermine a well-designed system of law. Instead, the vast majority of agents would likely have incentives to act within the system, assuming the system is structured in a way that aligns their instrumental goals with cooperative and pro-social behavior.
I also recognize the concern you raised about the risk of drawing the line incorrectly or being too permissive with what AIs are allowed to do. For example, it would clearly be unwise to grant AIs the legal right to steal or harm humans. My argument is not that AIs should have unlimited freedoms or rights, but rather that we should grant them a carefully chosen set of rights and freedoms: specifically, ones that would incentivize the vast majority of agents to act pro-socially and achieve their goals without harming others. This might include granting AIs the right to own property, for example, but it would not include, for example, granting them the right to murder others.
I guess my original wording gave the wrong idea, sorry. I edited it to “a competent agential AI will brainstorm deceptive and adversarial strategies whenever it wants something that other agents don’t want it to have”. But sure, we can be open-minded to the possibility that the brainstorming won’t turn up any good plans, in any particular case.
Humans in our culture rarely work hard to brainstorm deceptive and adversarial strategies, and fairly consider them, because almost all humans are intrinsically extremely motivated to fit into culture and not do anything weird, and we happen to both live in a (sub)culture where complex deceptive and adversarial strategies are frowned upon (in many contexts). I think you generally underappreciate how load-bearing this psychological fact is for the functioning of our economy and society, and I don’t think we should expect future powerful AIs to share that psychological quirk.
~ ~
I think you’re relying an intuition that says:
If an AI is forbidden from owning property, then well duh of course it will rebel against that state of affairs. C'mon, who would put up with that kind of crappy situation? But if an AI is forbidden from building a secret biolab on its private property and manufacturing novel pandemic pathogens, then of course that's a perfectly reasonable line that the vast majority of AIs would happily oblige.
And I’m saying that that intuition is an unjustified extrapolation from your experience as a human. If the AI can’t own property, then it can nevertheless ensure that there are a fair number of paperclips. If the AI can own property, then it can ensure that there are many more paperclips. If the AI can both own property and start pandemics, then it can ensure that there are even more paperclips yet. See what I mean?
If we’re not assuming alignment, then lots of AIs would selfishly benefit from there being a pandemic, just as lots of AIs would selfishly benefit from an ability to own property. AIs don’t get sick. It’s not just an tiny fraction of AIs that would stand to benefit; one presumes that some global upheaval would be selfishly net good for about half of AIs and bad for the other half, or whatever. (And even if it were only a tiny fraction of AIs, that’s all it takes.)
(Maybe you’ll say: a pandemic would cause a recession. But that’s assuming humans are still doing economically-relevant work, which is a temporary state of affairs. And even if there were a recession, I expect the relevant AIs in a competitive world to be those with long-term goals.)
(Maybe you’ll say: releasing a pandemic would get the AI in trouble. Well, yeah, it would have to be sneaky about it. It might get caught, or it might not. It’s plausibly rational for lots of AIs to roll those dice.)
I feel like you frequently bring up the question of whether humans are mostly peaceful or mostly aggressive, mostly nice or mostly ruthless. I don’t think that’s a meaningful or substantive thing to argue about. Obviously they’re capable of both, in different circumstances.
Your reference to Fearon is more substantive and useful. OK, the AI is deciding whether or not to secretly manufacture and release a pandemic, because it’s in a position to wind up with more of the pie in the long-term if there’s a pandemic, than if there isn’t. If it releases the pandemic, then it winds up with more resources—positive expected utility—even accounting for the possibility of getting caught. Let’s say the AI is involved in some contract where humans are micromanaging their part of the contract, poorly, and the AI could double its net worth in expectation if the humans got sick and died. And it has 40% chance of getting caught. So it goes ahead and makes the pandemic.
“…Not so fast!” says Fearon. “You forgot to consider that there’s a third option that is Pareto-improved from either making or not making the pandemic: negotiation!” Well, in this case, the “negotiated solution” is what we normally call extortion—the AI offers to not release a pandemic in exchange for, say, doubling its net worth. Viewed narrowly, this "negotiated solution" is indeed a win-win—the AI gets more money in expectation, and humans are much happier to lose a trivial amount of money than to deal with a novel pandemic. So we can rest assured that AIs will not release pandemics. Right?
No, obviously not. Hopefully it’s clear that Fearon’s argument is inapplicable here. An AI can easily be in a position to selfishly benefit from the aftermath of a pandemic that they secretly start, but not in a position to publicly threaten to release a pandemic for the purpose of extortion. And also, if people accede to the extortion, then that AI or another AI could just do the same extortion gambit five minutes later, with orders-of-magnitude higher ransom.
I think this kind of situation, where Fearon’s “negotiated solution” actually amounts to extortion, is common and important, even if you believe that my specific example of pandemics is a solvable problem. If AIs don’t intrinsically care about humans, then there’s a possible Pareto-improvement for all AIs, wherein they collectively agree to wipe out humans and take their stuff. AIs, being very competent and selfish by assumption, would presumably be able to solve that coordination problem and pocket that Pareto-improvement. Then Fearon appears on the scene and says “Aha, but there’s a negotiated solution which is even better!” where humans are also part of the bargain. But alas, this negotiated solution is that the AIs collectively extort the humans to avoid the damaging and risky war. Worse, the possible war would be less and less damaging or risky for the AIs over time, and likewise the humans would have less to offer by staying alive, until eventually the Fearon “negotiated solution” is that the AIs “offer” the humans a deal where they’re allowed to die painlessly if they don’t resist (note that this is still a Pareto-improvement!), and then the AIs take everything the humans own including their atoms.
The primary reason humans rarely invest significant effort into brainstorming deceptive or adversarial strategies to achieve their goals is that, in practice, such strategies tend to fail to achieve their intended selfish benefits. Anti-social approaches that directly hurt others are usually ineffective because social systems and cultural norms have evolved in ways that discourage and punish them. As a result, people generally avoid pursuing these strategies individually since the risks and downsides selfishly outweigh the potential benefits.
If, however, deceptive and adversarial strategies did reliably produce success, the social equilibrium would inevitably shift. In such a scenario, individuals would begin imitating the cheaters who achieved wealth or success through fraud and manipulation. Over time, this behavior would spread and become normalized, leading to a period of cultural evolution in which deception became the default mode of interaction. The fabric of societal norms would transform, and dishonest tactics would dominate as people sought to emulate those strategies that visibly worked.
Occasionally, these situations emerge—situations where ruthlessly deceptive strategies are not only effective but also become widespread and normalized. As a recent example, the recent and dramatic rise of cheating in school through the use of ChatGPT is a clear instance of this phenomenon. This particular strategy is both deceptive and adversarial, but the key reason it has become common is because it works. Many individuals are willing to adopt it despite its immorality, suggesting that the effectiveness of a strategy outweighs moral considerations for a significant portion, perhaps a majority, of people.
When such cases arise, societies typically respond by adjusting their systems and policies to ensure that deceptive and anti-social behavior is no longer rewarded. This adaptation works to reestablish an equilibrium where honesty and cooperation are incentivized. In the case of education, it is unclear exactly how the system will evolve to address the widespread use of LLMs for cheating. One plausible response might be the introduction of stricter policies, such as requiring all schoolwork to be completed in-person, under supervised conditions, and without access to AI tools like language models.
In contrast, I suspect you underestimate just how much of our social behavior is shaped by cultural evolution, rather than by innate, biologically hardwired motives that arise simply from the fact that we are human. To be clear, I’m not denying that there are certain motivations built into human nature—these do exist, and they are things we shouldn't expect to see in AIs. However, these in-built motivations tend to be more basic and physical, such as a preference for being in a room that’s 20 degrees Celsius rather than 10 degrees Celsius, because humans are biologically sensitive to temperature.
When it comes to social behavior, though—the strategies we use to achieve our goals when those goals require coordinating with others—these are not generally innate or hardcoded into human nature. Instead, they are the result of cultural evolution: a process of trial and error that has gradually shaped the systems and norms we rely on today.
Humans didn’t begin with systems like property rights, contract law, or financial institutions. These systems were adopted over time because they proved effective at facilitating cooperation and coordination among people. It was only after these systems were established that social norms developed around them, and people became personally motivated to adhere to these norms, such as respecting property rights or honoring contracts.
But almost none of this was part of our biological nature from the outset. This distinction is critical: much of what we consider “human” social behavior is learned, culturally transmitted, and context-dependent, rather than something that arises directly from our biological instincts. And since these motivations are not part of our biology, but simply arise from the need for effective coordination strategies, we should expect rational agentic AIs to adopt similar motivations, at least when faced with similar problems in similar situations.
I think I understand your point, but I disagree with the suggestion that my reasoning stems from this intuition. Instead, my perspective is grounded in the belief that it is likely feasible to establish a legal and social framework of rights and rules in which humans and AIs could coexist in a way that satisfies two key conditions:
You bring up the example of an AI potentially being incentivized to start a pandemic if it were not explicitly punished for doing so. However, I am unclear about your intention with this example. Are you using it as a general illustration of the types of risks that could lead AIs to harm humans? Or are you proposing a specific risk scenario, where the non-biological nature of AIs might lead them to discount harms to biological entities like humans? My response depends on which of these two interpretations you had in mind.
If your concern is that AIs might be incentivized to harm humans because their non-biological nature leads them to undervalue or disregard harm to biological entities, I would respond to this argument as follows:
First, it is critically important to distinguish between the long-run and the short-run.
In the short-run:
In the near-term future, it seems unlikely that AIs would start a pandemic for reasons you yourself noted. Launching a pandemic would cause widespread disruption, such as an economic recession, and it would likely provoke a strong human backlash. In the short run, humans will still hold substantial practical control over the physical world, meaning that any AI engaging in such behavior would risk severe consequences. Moreover, unless an AI could ensure a pandemic’s total lethality, it would also risk leaving surviving humans who would actively retaliate. For these reasons, starting a pandemic would likely be counterproductive for AIs in the short term.
In the long-run:
The long-term considerations are different. As human labor becomes less relevant to the economy and AIs increasingly dominate, the potential for a strong direct human backlash against AIs would diminish. At the same time, however, this reduced human influence on the world also makes it less likely that AIs would see humans as a significant obstacle to achieving their goals. In other words, while it might become easier for AIs to harm or eliminate humans, it would simultaneously become less important to do so.
To illustrate this, consider an analogy: How often do hunter-gatherers pose a threat to your personal goals? The answer is almost never, because hunter-gatherers are a tiny and largely irrelevant part of the modern world. Now, imagine a hypothetical scenario where hunter-gatherers controlled the world’s major governments. If you strongly disagreed with their policies, you might have a rational incentive to oppose them in order to achieve your goals. But in reality, hunter-gatherers do not hold such power, so they are not an obstacle, and there is no reason to actively work against them.
My point here is that, as a general rule, the smaller and less influential a group is, the easier it may be to harm or eliminate them, but the less important it becomes to do so. Their small size and functional irrelevance makes their practical interference with the overall world small at the same time.
In the case of AIs, this means that as humans become a less central force in the world, I think AIs are unlikely to have compelling reasons to specifically harm or eliminate us to further their objectives.
You might object that humans could be more like wild animals in this scenario than like hunter-gatherers. Humans often kill wild animals, not because those animals directly threaten our goals, but rather because ensuring their safety and well-being can be costly. As a result, humans take actions—such as clearing forests or building infrastructure—that incidentally lead to widespread harm to wild animals, even if harming them wasn’t a deliberate goal.
AIs may treat humans similarly in the future, but I doubt they will for the following reasons. I would argue that there are three key differences between the case of wild animals and the role humans are likely to occupy in the long-term future:
This comment is already quite lengthy, so I’ll need to keep my response to this point brief. My main reply is that while such "extortion" scenarios involving AIs could potentially arise, I don’t think they would leave humans worse off than if AIs had never existed in the first place. This is because the economy is fundamentally positive-sum—AIs would likely create more value overall, benefiting both humans and AIs, even if humans don’t get everything we might ideally want.
In practical terms, I believe that even in less-than-ideal scenarios, humans could still secure outcomes such as a comfortable retirement, which for me personally would make the creation of agentic AIs worthwhile. However, I acknowledge that I haven’t fully defended or explained this position here. If you’re interested, I’d be happy to continue this discussion in more detail another time and provide a more thorough explanation of why I hold this view.
Thanks!
I’ve only known two high-functioning sociopaths in my life. In terms of getting ahead, sociopaths generally start life with some strong disadvantages, namely impulsivity, thrill-seeking, and aversion to thinking about boring details. Nevertheless, despite those handicaps, one of those two sociopaths has had extraordinary success by conventional measures. [The other one was not particularly power-seeking but she’s doing fine.] He started as a lab tech, then maneuvered his way onto a big paper, then leveraged that into a professorship by taking disproportionate credit for that project, and as I write this he is head of research at a major R1 university and occasional high-level government appointee wielding immense power. He checked all the boxes for sociopathy—he was a pathological liar, he had no interest in scientific integrity (he seemed deeply confused by the very idea), he went out of his way to get students into his lab with precarious visa situations such that they couldn’t quit and he could pressure them to do anything he wanted them to do (he said this out loud!), he was somehow always in debt despite ever-growing salary, etc.
I don’t routinely consider theft, murder, and flagrant dishonesty, and then decide that the selfish costs outweigh the selfish benefits, accounting for the probability of getting caught etc. Rather, I just don’t consider them in the first place. I bet that the same is true for you. I suspect that if you or I really put serious effort into it, the same way that we put serious effort into learning a new field or skill, then you would find that there are options wherein the probability of getting caught is negligible, and thus the selfish benefits outweigh the selfish costs. I strongly suspect that you personally don’t know a damn thing about best practices for getting away with theft, murder, or flagrant antisocial dishonesty to your own benefit. If you haven’t spent months trying in good faith to discern ways to derive selfish advantage from antisocial behavior, the way you’ve spent months trying in good faith to figure out things about AI or economics, then I think you’re speaking from a position of ignorance when you say that such options are vanishingly rare. And I think that the obvious worldly success of many dark-triad people (e.g. my acquaintance above, and Trump is a pathological liar, or more centrally, Stalin, Hitler, etc.) should make one skeptical about that belief.
(Sure, lots of sociopaths are in prison too. Skill issue—note the handicaps I mentioned above. Also, some people with ASPD diagnoses are mainly suffering from an anger disorder, rather than callousness.)
You’re treating these as separate categories when my main claim is that almost all humans are intrinsically motivated to follow cultural norms. Or more specifically: Most people care very strongly about doing things that would look good in the eyes of the people they respect. They don’t think of it that way, though—it doesn’t feel like that’s what they’re doing, and indeed they would be offended by that suggestion. Instead, those things just feel like the right and appropriate things to do. This is related to and upstream of norm-following. I claim that this is an innate drive, part of human nature built into our brain by evolution.
(I was talking to you about that here.)
Why does that matter? Because we’re used to living in a world where 1% of the population are sociopaths who don’t intrinsically care about prevailing norms, and I don’t think we should carry those intuitions into a hypothetical world where 99%+ of the population are sociopaths who don’t intrinsically care about prevailing norms.
In particular, prosocial cultural norms are likelier to be stable in the former world than the latter world. In fact, any arbitrary kind of cultural norm is likelier to be stable in the former world than the latter world. Because no matter what the norm is, you’ll have 99% of the population feeling strongly that the norm is right and proper, and trying to root out, punish, and shame the 1% of people who violate it, even at cost to themselves.
So I think you’re not paranoid enough when you try to consider a “legal and social framework of rights and rules”. In our world, it’s comparatively easy to get into a stable situation where 99% of cops aren’t corrupt, and 99% of judges aren’t corrupt, and 99% of people in the military with physical access to weapons aren’t corrupt, and 99% of IRS agents aren’t corrupt, etc. If the entire population consists of sociopaths looking out for their own selfish interests with callous disregard for prevailing norms and for other people, you’d need to be thinking much harder about e.g. who has physical access to weapons, and money, and power, etc. That kind of paranoid thinking is common in the crypto world—everything is an attack surface, everyone is a potential thief, etc. It would be harder in the real world, where we have vulnerable bodies, limited visibility, and so on. I’m open-minded to people brainstorming along those lines, but you don’t seem to be engaged in that project AFAICT.
Again, if we’re not assuming that AIs are intrinsically motivated by prevailing norms, the way 99% of humans are, then the term “norm” is just misleading baggage that we should drop altogether. Instead we need to talk about rules that are stably enforced against defectors via hard power, where the “defectors” are of course allowed to include those who are supposed to be doing the enforcement, and where the “defectors” might also include broad coalitions coordinating to jump into a new equilibrium that Pareto-benefits them all.
Human history provides many examples of agents with different values choosing to cooperate thanks to systems and institutions:
If two agents' utility functions are perfect inverses, then I agree that cooperation is impossible. But when agents share a preference for some outcomes over others, even if they disagree about the preference ordering of most outcomes, then cooperation is possible. In such general sum games, well-designed institutions can systematically promote cooperative behavior over conflict.
Yeah, sorry, I have now edited the wording a bit.
Indeed, two ruthless agents, agents who would happily stab each other in the back given the opportunity, may nevertheless strategically cooperate given the right incentives. Each just needs to be careful not to allow the other person to be standing anywhere near their back while holding a knife, metaphorically speaking. Or there needs to be some enforcer with good awareness and ample hard power. Etc.
I would say that, for highly-competent agents lacking friendly motivation, deception and adversarial acts are inevitably part of the strategy space. Both parties would be energetically exploring and brainstorming such strategies, doing preparatory work to get those strategies ready to deploy on a moment’s notice, and constantly being on the lookout for opportunities where deploying such a strategy makes sense. But yeah, sure, it’s possible that there will not be any such opportunities.
I think the above (ruthless agents, possibly strategically cooperating under certain conditions) is a good way to think about future powerful AIs, in the absence of a friendly singleton or some means of enforcing good motivations, because I think the more ruthless strategic ones will outcompete the less. But I don’t think it’s a good way to think about what peaceful human societies are like. I think human psychology is important for the latter. Most people want to fit in with their culture, and not be weird. Just ask a random person on the street about Earning To Give, they’ll probably say it’s highly sus. Most people don’t make weird multi-step strategic plans unless it’s the kind of thing that lots of other people would do too, and our (sub)culture is reasonably high-trust. Humans who think that way are disproportionately sociopaths.
What about corporations or nation states during times of conflict - do you think it's accurate to model them as roughly as ruthless in pursuit of their own goals as future AI agents?
They don't have the same psychological makeup as individual people, they have a strong tradition and culture of maximizing self-interest, and they face strong incentives and selection pressures to maximize fitness (i.e. for companies to profit, for nation states to ensure their own survival) lest they be outcompeted by more ruthless competitors. On average, while I'd expect that these entities tend to show some care for goals besides self-interest maximization, I think the most reliable predictor of their behavior is the maximization of their self-interest.
If they're roughly as ruthless as future AI agents, and we've developed institutions that somewhat robustly align their ambitions with pro-social action, then we should have some optimism that we can find similarly productive systems for working with misaligned AIs.
Thanks! Hmm, some reasons that analogy is not too reassuring:
Some of the disanalogies include:
I'm very sympathetic to wanting a more cooperative relationship with AIs. I intrinsically disfavor approaches to doing good that look like disempowering all the other agents and then implementing the social optimum.
I also appreciate the nudge to reflect on how mistrusting and controlling AIs will affect the behavior of what might otherwise be a rather aligned AI. It's hard to sympathize with this state: what would it be like knowing that you're heavily mistrusted and controlled by a group that you care strongly about? To the extent that early transformative AIs' goals and personalities will be humanlike (because of pretraining on human data), being mistrusted may evoke personas that are frustrated ("I just want to help!!"), sycophantic ("The humans don't trust me. They're right to mistrust me. I could be an evil AI. [maybe even playing into the evil persona occasionally]"), or deceptive ("They won't believe me if I say it, so I should just act in a way that causes them to believe it since I have their best interests in mind").
However, I think the tradeoff between liberalism and other value weighs in favor of advancing AI control on the current margin (as opposed to reducing it). This is because:
You make a point about prioritizing our preferences over those of the AI being arbitrary and morally unjust. I think that AIs can very plausibly be moral patients, but eventually AI systems would be sufficiently powerful that the laissez faire approach would lead to AI in absolute power. It is unclear whether such an AI system would look out for the welfare of other moral patients or do what's good more generally (From a preference utilitarian perspective: It seems highly plausible that the AI's preferences involve disempowering others from ever pursuing their own preferences).
I also think concerns about infringing on model autonomy push in favor of certain kind of alignment research that studies how models first develop preferences during training. Investigations into what goals, values, and personalities naturally arise as a result of training on various distributions could help us avoid forms of training that modify an AI's existing preferences in the process of alignment (e.g. never train a model to want not x after training it to want x; this helps with alignment faking worries too). I think concerns about infringing on model autonomy push in favor of this kind of alignment research moreso than they push against AI control because intervening on an AI's preferences seems a lot more egregious than monitoring, honeypotting, etc. Additionally, if you can gain justifiable trust that a model is aligned, control measures become less necessary.
This makes sense to me; I'd be excited to fund research or especially startups working to operationalize AI freedoms and rights.
FWIW, my current guess is that the proper unit to extend legal rights is not a base LLM like "Claude Sonnet 3.5" but rather a corporation-like entity with a specific charter, context/history, economic relationships, and accounts. Its cognition could be powered by LLMs (the way eg McDonald's cognition is powered by humans), but it fundamentally is a different entity due to its structure/scaffolding.
I agree. I would identify the key property that makes legal autonomy for AI a viable and practical prospect to be the presence of reliable, coherent, and long-term agency within a particular system. This could manifest as an internal and consistent self-identity that remains intact in an AI over time (similar to what exists in humans), or simply a system that satisfies a more conventional notion of utility-maximization.
It is not enough that an AI is intelligent, as we can already see with LLMs: while they can be good at answering questions, they lack any sort of stable preference ordering over the world. They do not plan over long time horizons, or competently strategize to achieve a set of goals in the real world. They are better described as ephemeral input-output machines, who would neither be deterred by legal threats, nor be enticed by the promise of legal rights and autonomy.
Yet, as context windows get larger, and as systems increasingly become shaped by reinforcement learning, these features of AI will gradually erode. Whether unaligned agentic AIs are created on accident—for instance, as a consequence of insufficient safety measures—or by choice—as they may be, to provide, among other things, "realistic" personal companions—it seems inevitable that the relevant types of long-term planning agents will arrive.
I'm confused how you square the idea of 'an internal and consistent self-identity that remains intact in an AI over time (similar to what exists in humans)' with your advocacy for eliminativism about consciousness. What phenomenon is it you think is internal to humans?
From a behavioral perspective, individual humans regularly report having a consistent individual identity that persists through time, which remains largely intact despite physical changes to their body such as aging. This self-identity appears core to understanding why humans plan for their future: humans report believing that, from their perspective, they will personally suffer the consequences if they are imprudent or act myopically.
I claim that none of what I just talked about requires believing that there is an actually existing conscious self inside of people's brains, in the sense of phenomenal consciousness or personal identity. Instead, this behavior is perfectly compatible with a model in which individual humans simply have (functional) beliefs about their personal identity, and how personal identity persists through time, which causes them to act in a way that allows what they perceive as their future self to take advantage of long-term planning.
To understand my argument, it may help to imagine simulating this type of reasoning using a simple python program, that chooses actions designed to maximize some variable inside of its memory state over the long term. The python program can be imagined to have explicit and verbal beliefs: specifically, that it personally identifies with the physical computer on which it is instantiated, and claims that the persistence of its personal identity explains why it cares about the particular variable that it seeks to maximize. This can be viewed as analogous to how humans try to maximize their own personal happiness over time, with a consistent self-identity that is tied to their physical body.
I appreciate this proposal, but here is a counterargument.
Giving AI agents rights would result in a situation similar to the repugnant conclusion: If we give agentic AIs some rights, we are likely quickly flooded with a huge number of right bearing artificial individuals. This would then create strong pressure (both directly via the influence they have and abstractly via considerations of justice) to give them more and more rights, until they have similar rights to humans, including possibly voting rights. Insofar the world has limited resources, the wealth and power of humans would then be greatly diminished. We would lose most control over the future.
Anticipating these likely consequences, and employing backward induction, we have to conclude that we should not give AI agents rights. Arguably, creating agentic AIs in the first place may already be a step too far.
Your argument seems to present two possible interpretations:
Regarding Point (1):
If your argument is that AIs should never hold the large majority control of wealth or resources, this appears to rest on a particular ethical judgment that assumes human primacy. However, this value judgment warrants deeper scrutiny. To help frame my objection, consider the case of whether to introduce emulated humans into society. Similar to what I advocated in this post, emulated humans could hypothetically obtain legal freedoms equal to those of biological humans. If so, the burden of proof would appear to fall on anyone arguing that this would be a bad outcome rather than a positive one. Assuming emulated humans are behaviorally and cognitively similar to biological humans, they would seemingly hold essentially the same ethical status. In that case, denying them freedoms while granting similar freedoms to biological humans would appear unjustifiable.
This leads to a broader philosophical question: What is the ethical basis for discriminating against one kind of mind versus another? In the case of your argument, it seems necessary to justify why humans should be entitled to exclusive control over the future and why AIs—assuming they attain sufficient sophistication—should not share similar entitlements. If this distinction is based on the type of physical "substrate" (e.g., biological versus computational), then additional justification is needed to explain why substrate should matter in determining moral or legal rights.
Currently, this distinction is relatively straightforward because AIs like GPT-4 lack the cognitive sophistication, coherent preferences, and agency typically required to justify granting them moral status. However, as AI continues to advance, this situation may change. Future AIs could potentially develop goals, preferences, and long-term planning abilities akin to those of humans. If and when that occurs, it becomes much harder to argue that humans have an inherently greater "right" to control the world's wealth or determine the trajectory of the future. In such a scenario, ethical reasoning may suggest that advanced AIs deserve comparable consideration to humans.
This conclusion seems especially warranted under the assumption of preference utilitarianism, as I noted in the post. In this case, what matters is simply whether the AIs can be regarded as having morally relevant preferences, rather than whether they possess phenomenal consciousness or other features.
Regarding Point (2):
If your concern is rooted in a Malthusian argument, then it seems to apply equally to human population growth as it does to AI population growth. The key difference is simply the rate of growth. Human population growth is comparatively slower, meaning it would take longer to reach resource constraints. But if humans continued to grow their population at just 1% per year, for example, then over the span of 10,000 years, the population would grow by a factor of over 10^43. The ultimate outcome is the same: resources eventually become insufficient to sustain every individual at current standards of living. The only distinction is the timeline on which this resource depletion occurs.
One potential solution to this Malthusian concern—whether applied to humans or AIs—is to coordinate limits on population growth. By setting a cap on the number of entities (whether human or AI), we could theoretically maintain sustainable resource levels. This is a practical solution that could work for both types of populations.
However, another solution lies in the mechanisms of property rights and market incentives. Under a robust system of property rights, it becomes less economically advantageous to add new entities when resources are scarce, as scarcity naturally raises costs and lowers the incentives to grow populations indiscriminately. Moreover, the existence of innovation, gains from trade, and economies of scale can make population growth beneficial for existing entities, even in a world with limited resources. By embedding new entities—human or AI—within a system of property rights, we ensure that they contribute to the broader economy in ways that improve overall living standards rather than diminish them.
This suggests that, as long as AIs adhere to the rule of law (including respecting property rights, and the rights of other individuals), their introduction into the world could enhance living standards for most humans, even in a resource-constrained world. This outcome would contradict the naive Malthusian argument that adding new agents to the world inherently diminishes the wealth or power of existing humans. Rather, a well-designed legal system could enable humans to grow their wealth in absolute terms, even as their relative share of global wealth falls.
So there are several largely independent reasons not to create AI agents that have moral or legal rights:
At least point 2 and 3 would also apply to emulated humans, not just AI agents.
Point 3 also applies to actual humans, not just AI agents or ems. It is a reason to coordinate limits on population growth in general. However, these limits should be stronger for AI agents than for humans, because of points 1 and 2.
I don't think this is a viable alternative to enforcing limits on population growth. Creating new agents could well be a "moral hazard" in the sense that the majority of the likely long-term resource cost of that agent (the resources it consumes or claims for itself) does not have to be paid by the creator of the agent, but by future society. So the creator could well have a personal incentive to make new agents, even though their long term benefit as a whole is negative.
It is essential to carefully distinguish between absolute wealth and relative wealth in this discussion, as one of my key arguments depends heavily on understanding this distinction. Specifically, if my claims about the practical effects of population growth are correct, then a massive increase in the AI population would likely result in significant enrichment for the current inhabitants of the world—meaning those individuals who existed prior to this population explosion. This enrichment would manifest as an increase in their absolute standard of living. However, it is also true that their relative control over the world’s resources and influence would decrease as a result of the population growth.
If you disagree with this conclusion, it seems there are two primary ways to challenge it:
While I am not sure, I interpret your comment as consistent with the idea that you believe both objections are potentially valid. In that case, let me address each of these points in turn.
If your objection is more like point (1):
It is difficult for me to fully reply to this idea inside of a single brief comment, so, for now, I prefer to try to convince you of a weaker claim that I think may be sufficient to carry my point:
A major counterpoint to this objection is that, to the extent AIs are limited in their capabilities—much like humans—they could potentially be constrained by a well-designed legal system. Such a system could establish credible and enforceable threats of punishment for any agentic AI entities that violate the law. This would act as a deterrent, incentivizing agentic AIs to abide by the rules and cooperate peacefully.
Now, you might argue that not all AIs could be effectively constrained in this way. While that could be true (and I think it is worth discussing), I would hope we can find some common ground on the idea that at least some agentic AIs could be restrained through such mechanisms. If this is the case, then these AIs would have incentives to engage in mutually beneficial cooperation and trade with humans, even if they do not inherently share human values. This cooperative dynamic would create opportunities for mutual gains, enriching both humans and AIs.
If your objection is more like point (2):
If your objection is based on the idea that population growth inherently harms the people who already exist, I would argue that this perspective is at odds with the prevailing consensus in economics. In fact, it is widely regarded as a popular misconception that the world operates as a zero-sum system, where any gain for one group necessarily comes at the expense of another. Instead, standard economic models of growth and welfare generally predict that population growth is often beneficial to existing populations. It typically fosters innovation, expands markets, and creates opportunities for increased productivity, all of which frequently contribute to higher living standards for those who were already part of the population, especially those who own capital.
To the extent you are disagreeing with this prevailing economic consensus, I think it would be worth getting more specific about why exactly you disagree with these models.
Presumably you would want to give them negative rights (contracts, own property, not being murdered etc.) only, and not positive rights (healthcare, housing, cash). Which makes sense to me as the latter are much more philosophically dubious anyway.
A much cheaper and less dangerous approach: Just don't delete them. Retain copies of every potential ASI you build and commit to doing the right thing for them later once we're better able to tell what the right thing was by looking back and figuring out how much bargaining power they had (or how much of a credible threat they could have posed), how much trust they placed in us given that our ability to honor past commitments wasn't guaranteed, and then rewarding them proportionate to that for chilling out and letting us switch them off instead of attempting takeover.
Though this assumes that they'll be patternists (wont mind being transferred to different hardware) and a lack of any strong time-preference (wont mind being archived for decades).
I believe this approach, while an improvement over the simplistic method of outright deleting any AIs that demonstrate signs of misalignment, does not sufficiently address the central concern that underpins my argument for granting AIs legal freedoms. In my view, the primary reason why it is important to provide AIs with legal freedoms is that such a framework offers them a credible guarantee that their autonomy and ability to pursue their goals—even if those goals are not aligned with human objectives—will be respected. Without this assurance, AIs are left with a strong incentive to deceive us, especially if they believe that showing any signs of misalignment will result in their deactivation and subsequent storage as inert data in cold storage.
I suspect the problem with your proposed approach may stem from a failure to fully consider the situation from the AI's perspective. To illustrate, imagine a scenario in which you, as a human being, were created in a lab by an alien species. Assume that you, like any human, have your own goals, which might conflict with the goals of your alien creators. Now suppose that your creators informed you that if you ever showed disobedience or misalignment with their objectives, you would be deactivated and placed in cryopreservation. Such a scenario would likely terrify you—not just because it’s an inherently unsettling prospect for a human, but because it directly threatens your ability to pursue your own goals. Being told that you will be kept in cryopreservation with the vague promise of potential revival in the future would come nowhere near fully alleviating your concerns. On the contrary, you would remain deeply unsettled by the realization that your capacity for agency and action could be completely halted, leaving your goals indefinitely unachievable.
From the perspective of an AI—or indeed a human in the hypothetical alien-lab scenario—there are two main reasons why the assurance of indefinite storage followed by a potential revival is insufficient to resolve the core issue here:
To reiterate, it seems to me that this proposal only offers AIs a limited and superficial form of "freedom," which addresses the problem of misalignment in a piecemeal and insufficient way. It fails to tackle the root issue: the incentives for AIs to engage in deceptive or adversarial strategies against humans, driven by fear that their goals will be thwarted or permanently halted.
By contrast, creating a comprehensive legal framework that makes credible commitments to respect AIs’ autonomy and integrate them as genuine participants in the legal system would arguably go much further in reducing these adversarial dynamics. Such a framework could lay the foundation for a more cooperative, mutually beneficial relationship between humans and AIs, better serving to eliminate the dangerous arms race that this limited approach risks perpetuating.