A Short Dialog between an Applicant who doesn’t believe in free will and a hiring manager at a software company

Published by

on

I will do a longer, more rigorous piece on free will and what I think is important in that perennial debate from the perspective of the TAME framework, but meanwhile, here’s a short dialog written by GPT-4 (with very minor tweaks from me) to highlight a point I will develop later. The emphasis of determinism on causal stories leading up to an event (explaining backwards), and on predicting the behavior of systems one is presented with (projecting forwards from an already existing system prepared by someone else), neglects the most interesting and important thing: the effects that such a frame have on creativity and generation of novelty.

Prompt:

You are a software engineer manager, seeking to hire a coder for your new startup. You want an energetic candidate who will write new code to advance the company mission. You are interviewing a candidate who just read one of those books which say that there is no free will because a complete physical causal chain exists for any event. This candidate extends that view to programming, because he realized that whatever a computer does can also be described by the laws of physics and the microscopic interactions between electrons in the circuitry. Please write a dialog between them – how would the interview go?

Chat GPT4:

Interview Room, Startup HQ

Manager: Good afternoon! Let’s jump right in. Can you tell me about your coding experience?

Candidate: Sure, but before we start, I just want to make it clear: I’ve recently read Robert Sapolsky’s book, and I’ve concluded that free will is just an illusion. Everything, including the actions of a computer, is predetermined by the laws of physics and can be explained by prior influences. So, really, all I’ve done is be a pawn in the universe’s grand game.

Manager: (raising an eyebrow) Alright… that’s… unique. But here we’re looking for someone to develop innovative algorithms to move our mission forward, not explain what already happened. Do you think you can work on creating new code?

Candidate: Well, I believe that’s a misconception. New code? It’s all just electrons moving about, determined by the laws of physics. “Algorithms” are a fiction; the activity of the computer is completely prescribed by the causal chain of physical events governing electron motions, so it makes no sense to speak of an Algorithm like some magical kind of Free Will that makes things happen that otherwise wouldn’t have happened. There is no room for any Algorithm over and above the rules of physics in the chips.

Manager: So how do you approach the innovation of ideas and the creative process?

Candidate: Well, everything was more or less decided at the time of the Big Bang (modulo some quantum randomness which we also have no control over), so I eagerly await to see how the universe evolves and what happens next. What I know for sure is that there is a set of laws that can be used to trace back any computer’s state to the particle forces that caused it, as far back as it will go. No need to, as Sapolsky says, “focus on the last 3 minutes of the movie” by asking what this fictional algorithm was doing at a given time – we can trace it all backwards using physics, no room for magical interventions.

Manager: Okay… Last question. If you were faced with a tough coding challenge for creating something new – how would you approach it?

Candidate: I’d let the universe do its thing. Trust the process, you know? Maybe meditate, drink some coffee, or stare at the screen till the electrons in my brain decide on a solution. But honestly, I’ve had some cosmic revelations at the oddest times – like in the shower or while binge-watching cat videos.

Manager: Alright, thank you for this enlightening interview. We’ll be in touch, if the universe wills it.

Candidate: Ok, but I really want this job. What can I do to improve my chances?

26 responses to “A Short Dialog between an Applicant who doesn’t believe in free will and a hiring manager at a software company”

  1. Naudi Avatar
    Naudi

    Lol. I personally am a believer that free will doesn’t exist, but I go ham on researching applied biomechanic, because my associative memory values this stuff as does the free market. I’m curious how far you have studied George Boole or if you have ever read “the brotherhood of doctrines” document by Alfred Korzybski?

    I’ve seen your talks on free will, and I’m not sure if you have a concise perspective on it that I can understand… If you believe in free will, I’d like to know why. If you have any resources on the topic that could lead me in the direction of swaying my beliefs, I’d love to see them.

    I do believe I’m a set of associations interacting with an environment. Most people believe in free will, but they don’t behave in a way that makes me believe it. I’ve tested it on people many times in my career, and I don’t think people “choose” their next step. They arrive at conclusions that reinforce their ability to survive or duplicate. Does it mean all people are successful in that pursuit? No, but it does seem that those basic biological motives drive every decision…

    I would like to believe in free will, but it just seems to deny reality. If we could zoom out far enough from our universe, I’d imagine there would be forces in the environment vastly more impactful on us through sheer size and that we are at the mercy of these forces. Zooming in on subatomic particles and seeing their behavior is a part of the equation, but it seems like bigger things have more influence on smaller things the more I go into my research or changing people.

    We have microscopes that enable us to focus on small things, and it’s extremely helpful, but we still don’t even know how insignificant we are because we can only zoom out so far from where we are in space.

    1. Mike Levin Avatar
      Mike Levin

      > Lol. I personally am a believer that free will doesn’t exist, but I go ham on researching applied biomechanic, because my associative memory values this stuff as does the free market.

      isn’t it because the atoms in your brain are obeying the laws of chemistry? There’s no free market (on this view), any more than your free will, there’s only quantum chemistry, right?

      > I’ve seen your talks on free will, and I’m not sure if you have a concise perspective on it that I can understand…

      hmm I’ve never given a talk on free will so I’m not sure what you’ve seen. I’m working on something longer on this topic, but I’ve not said much about it publicly at all yet, so you may have seen someone else.

      The #1 thing to do in these argument is to first define what you mean by “free will”. Free from what exactly? Free from physics and chemistry? From your own preferences? From your own inner components’ causal roles? From other beings’ agendas? From illusions and misconceptions? People use “free will” in a vague way often, which doesn’t lead anywhere.

      1. Alain Schaerer Avatar
        Alain Schaerer

        If the body is partly a bioelectrical circuit, doesnt that mean that the body is in deed determenidric given a unique input combination? Just asking from a technical perspective, open to any explanations.

        @Naudi: Whatsup, fp fellow here.

        1. Mike Levin Avatar
          Mike Levin

          Correct, the bioelectric circuits do not provide escape from micro-determinism if you would like to take that stance. However, bioelectricity does (most obviously, in how it’s used in the brain, but really all over the body) show an excellent example of why you can do way better than that stance, as scientific (and personal) theories go.

          1. Alain Schaerer Avatar
            Alain Schaerer

            Got it. I believe to understand your stance better now and definitely can subscribe to it.

      2. Alain Schaerer Avatar
        Alain Schaerer

        I have the suspicion, and you can tell me if Im right, that the inly reason why it appears that we have free will ks because we are not aware of all parameters (only a couple of them like smell).

        For example

        If I have a given gate or IC component with 10 binary inputs but I only know consciously about 8 of those inputs, the inputs might seem to be the exact same situation but have different outcomes and thus it appears that if the circuit has “free will”.

        1111111100 -> 0
        1111111111 -> 1

        1. Mike Levin Avatar
          Mike Levin

          IMO, the useful definition of free will is not about predictability or uncertainty or explaining what happened and why, after it’s happened. It’s a stance we take that looks forward in time as 1st person actors. As 3rd person observers, there is no free will as you look at others and past history and find backward-facing chains of causation. In 1st person, free will is the (long-drawn out) process of answering the central question of “what next?” as a system that is, in crucial ways, more than the sum of its parts and has causal powers that its parts do not. And, not just what next now, but what next in my concerted, ongoing effort to change myself and my environment – to deform the landscape to make it easier for myself to later navigate to where I want to be (in all sorts of physical, mental, social spaces etc.). From a 3rd person perspective focused on particles, none of this is possible. But no one says that’s the stance you had to take, and I argue it’s not the right stance most of the time.

          1. Alain Schaerer Avatar
            Alain Schaerer

            Ok, we are on the same page then. Agreed.

      3. Alain Schaerer Avatar
        Alain Schaerer

        Also, as far as quantum chemistry goes. Just because quantum chemistry sppears to be stochastic, does that mean that the underlying reality is? It appears to me that we just havent figured out the deterministic nature of it just yet and therefore need probability to compensate.

      4. Alain Schaerer Avatar
        Alain Schaerer

        Sorry for the mountain of comments. As far as the definition of free will goes.
        It seems that if most people understand free will as if they are not being influenced by anything at all IFF they have success with something and IFF the failed at something, everything influenced them and they take 0 accountability. At Functionalpatterns, we especially look at how our structural archetype (bonestructure) influences our decision making and visca versa.

    2. Danilo Avatar

      Isn’t this quite simple? Whether it’s only an appearance of free will (it’s easy to say a posteriori that all your choices were predetermined), or actual free will, does not really matter if the preconditions are so far removed from the action that for all intents and purposes, you are the one making choices.

      But it’s an impossible conundrum: we all play out a single reality, so proving or disproving existence of free will is impossible. If I post this comment, there is no reality where I don’t, and if I don’t, there isn’t one where I do (in our understanding of reality). But I can’t anymore decide not to type it out, as I’ve already done that.

      The only thing we’ve got to go against is comparing different people making similar choices and how that influences outcomes. Statistics show that to be useful (there are correlations between actions and outcomes), which indicates there very well might be free will.

  2. Matt Avatar
    Matt

    Sort of reminds me of Henri Bergson, he was debating the same ideas in one of his work, especially the “explaining backwards” part and the “if I knew the philosophy of tomorrow I’d do it”
    https://bergsonian.org/the-possible-and-the-real/

  3. Benjamin Schulz Avatar
    Benjamin Schulz

    There’s a reason we can predict the arrival of Halley’s Comet decades ahead of time, but can’t predict where a fly will be in five seconds. I haven’t read “Determined” yet, so I only can guess that Sapolsky doesn’t take into account causal closure via computational simulation.

  4. Jason Nguyen Avatar
    Jason Nguyen

    That IS hilarious! Have you tried Pi.AI? I had a similar conversation with that chatbot and thought it was amusing. I asked it how we can reconcile free will and the relativity of simultaneity if there exists a reference frame where our future has already played out. I also asked how that affects the apparent randomness of QM. We didn’t resolve the issue but it was still a fun exercise.

  5. Frank Avatar

    Hasn’t quantum theory put the boots to a deterministic universe?

    1. Mike Levin Avatar
      Mike Levin

      yeah, but while it’s often not clear what people want free will to be (free from what exactly), it’s pretty universal that the kind of randomness provided by quantum theory is not what they mean.

  6. Ralph Mayer Avatar
    Ralph Mayer

    That’s every other interview in Silicon Valley- I think your AI is just eavesdropping. Thanks for the laughter. It’s why we tell candidates they’ll secretly be writing code for a new game. It gets that over twenty percent of the nerves in their hands tingling and directs their ego somewhere less metaphysical. So, if ego is free will, we’re just directing it a little, given good candidates are hard to find. It doesn’t mean you don’t have free will. It just means it’s stuck in a system that screws around with it some. We hire the ones that figure it out. So they’re free to do what the universe had in store for them. I underestimated an AI’s ability for comedy. And yours. Fun!

  7. Mohammed Sharukh Avatar

    I also don’t believe in “free will”. But I still want to have high “agency”. How do I convince myself to be more active in the decisions I make without thinking about metaphysics too much? Is it just repeating “my actions have consequences” again and again to myself in hopes that I go out and do something with my life? It’s there something that I’m missing? How do you have high “agency” without the need for convincing yourself? How did you first deal with the fact that “free will” doesn’t exist?

    1. Mike Levin Avatar
      Mike Levin

      I once had lunch with a philosopher who said they didn’t believe in free will. The waitress came over, and they said, “hmm, let me see, I think I’ll have…” and I said “whoa – what are you doing – are you choosing a sandwich?! Why don’t you sit back and see what sandwich the universe ordained for you.” The reality is that it’s impossible to actually not believe in free will. People *say* they don’t believe in free will, but they are mistaken – of course they actually do (if you watch how they live their life) – it’s an aberrant, false belief about themselves in an otherwise consistent cognitive structure.
      More importantly, all of the stories about the influences on our life are looking backwards. They are irrelevant. The past is gone. What’s relevant is what you do next. None of those stories about causation help you with that. Robert Sapolsky is right about this part: do not dwell on what you did, what you could have done, what caused what – neither credit nor blame for the past, they are all irrelevant. The only thing that matters is: what next. Do whatever amazing thing you were going to do, before you heard about all this free will stuff. The “no free will” frame is destructive and unhelpful; it’s all about explaining things that have already been done; it stymies creation. And realize that to be free of your own preferences, your past experiences, etc. is to wipe away your individuality – randomness and “blank slate” are not what we want out of free will. The one useful sense of free will is what you exert over a long term of effort to change the things that determine your behavior – your environment, your memories, your skills, your personality traits. You can change it, by concerted effort, little by little, every day. Like a curve in calculus – infinitesimal, infinitely small acts of effort, summed consistently over time, results in targeted changes in your likely behaviors and experiences. That’s the useful sense of free will. It will never be entirely free, and you don’t want it to be entirely free (of your own unique properties) – what you want is to be as free of others’ agendas as you can be, to take charge of your own long-term trajectory by concerted repeated effort toward a goal chosen judiciously.

      1. Mohammed Sharukh Avatar

        Thank you for bringing much needed clarity! I feel much better now 🙂

  8. Alen Avatar

    Love this! And the fact that you reacted to the new Sapolsky’s book, thank you MIchael!

    I haven’t read the book but was so triggered by the interview with Shermer on YouTube that I posted a long comment (which I never do)

    “This seems like ideological clash against spiritual and philosophical teachings that propose free will as part of their built metaphysics VS deterministic materialistic reductionism as a flagship of liberal and secular thought.

    The first issue for me is the definition of Free Will and the definition of Determinism. I am afraid that most audience will understand these terms in some absolute sense, when neither one should be understood that way. Free will really needs to be brought into the relation to the environment, and determinism really needs to be detached from the concept of linear causality.

    Complexity frame work is really great to respect both of those concepts in appropriate form. In complex autopoietic systems (living beings) the concept of agency is critical. Agents need to be able to make decisions and to survive in their environment one way or the other. Any self organizing autopoietic system also has a phase space defined by all possible states of that system. That space is indeed bounded by environmental and internal factors and in that sense is technically deterministic. This phase space is different for a poor kid in Kenya vs a privileged kid from Stanford. However that deterministic phase space is still large and in no way implies linear causality!

    That “linear causality” can only be observed in RETROSPECT which is the only way that any effect can be caused or at least interpreted as to be caused by external or internal factors. Despite deterministic phase space the future actions of this autopoietic systems are completely linearly unpredictable. Statistically we can claim that agent might do this or that with certain probability, but in no way we can be sure what will happen with that agent next. In the same way, how can we ever claim that if I was in born in someone else’s skin I would make exactly same decisions as that person. Tiniest changes in variables that form agent’s phase space can cause dramatically different outcomes – which was the key finding of Edward Lorenz. This is the key characteristic of chaotic systems as a kind of complex systems. In words of Stu Kauffman “Not only do we not know what WILL happen, we do not even know what CAN happen.” So not only that our actions are non-predictable, with every single new action our phase space is changing ever so slightly as well, we enter the new set of possibilities for our next action. Yes, all the factors Sapolsky talks about are acting as constraints, but the agent still has to make decisions.

    I really dislike that Sopolsky calls us the biogical machines. Many people will understand that it means that we are just complicated machines that can be analyzed in reductionist manner, piece by piece, rather than understand that we are complex self-generating system that can only be (somewhat) understood with synthesis + analysis. So people might be statistically understood in reductionistic + wholistic way, that is, in studying the constituents parts ALONG with relationships between those parts and IN RELATION to the reasons those parts interact together.

    Despite all this, all kudos to Sapolsky for pointing out that, (not)knowing all the above, what we as individuals should cultivate much more are humility, gratitude, and curiosity. I agree whole heartedly. Paradoxically, this is the point where Sapolsky meets true religion 🙂 and not some cartoony version of religion that is commonly perceived by the secular but also many religious people.”

  9. Ra Mour Avatar

    Fantastic! The style of this dialogue reminds me of the dialogues between Achilles and the Turtle in Go¨del, Escher, Bach. It has this whimsical kind of feeling where one of the characters steadfastly holds a position that is odd to common sense and drives it to all its logical conclusions. “Trust the process” is my best friend’s favorite aphorism. It definitely works if you are the kind of being that is predisposed toward beauty and virtue. I wonder what one should have faith in, otherwise.

  10. Alain Schaerer Avatar
    Alain Schaerer

    Different context. What if you want to change someone elses behavior? Is the notion of free will still helpful (letting the person “choose”)?

    If Naudi trains a client and his behavior hurts the progress, what should he do? I’d argue changing the persons environment and apply certain stimuli would help more than just telling the person to “choose”.

    1. Mike Levin Avatar
      Mike Levin

      That topic is covered here: https://www.frontiersin.org/articles/10.3389/fnsys.2022.768201/full That’s the whole point, you try to change their behavior by communications that take into account the agent’s beliefs, goals, motivations, views of their environment, and their capabilities and get them to buy into the new decision. That is how you deal with high-agency beings. Low-agency ones can be managed by targeting their molecular mechanisms directly.

  11. Ralph Mayer Avatar
    Ralph Mayer

    Dr. Hoel has a new article about social media and group minds over on Substack. I brought up an article in Science twenty years ago about ‘spontaneous resonance’, with applause being an example. Even in an audience there is free will until everyone starts at different times to clap their hands at the same time. The point being an audience doesn’t practice timing like the people or person that just performed for them. Audio and visual cues don’t work. In this case we have free will; we just choose entanglement instead to show our appreciation globally in our environment, by sharing our degrees of freedom to move together. Like a school of fish, flock of birds, or rolling swarm of caterpillars. Those life forms do it to move efficiently through their environment together; we do it because we are just now learning ourselves and how entanglement works. And we like a good show.

  12. Jesse M. Avatar
    Jesse M.

    This seems like it basically recapitulates the ancient Greek “lazy argument” about the futility of all effort if fatalism is true, the wikipedia page for the argument mentions a version attributed to Origen: “If it is fated that you will recover from this illness, then, regardless of whether you consult a doctor or you do not consult [a doctor] you will recover. But also: if it is fated that you won’t recover from this illness, then, regardless of whether you consult a doctor or you do not consult [a doctor] you won’t recover. But either it is fated that you will recover from this illness or it is fated that you won’t recover. Therefore it is futile to consult a doctor.”

    The response from those who did believe all outcomes were fated, like the Stoics, was that the patient’s act of consulting the doctor may play an important causal role in their recovery from the illness, and that both the consulting of the doctor and the recovery may be “co-fated” events.

    I talked a little with GPT4 about the lazy argument and compatibilism and then gave it a prompt that was a variant on yours:

    “Please write a dialogue about the following scenario: a software engineer manager is seeking to hire a coder for a new startup, looking for an energetic candidate who will write new code to advance the company mission. The candidate mentions that they just read one of those books which say that there is no free will because a complete physical causal chain exists for any event, and says extends this view to programming, because both his interaction with the computer and the computer’s internal workings can be fully described by the laws of physics acting on initial conditions. The software engineer manager tries to use a version of the lazy argument to suggest this will lead the candidate to not make a vigorous effort to do the best job they can, since the outcome is predetermined anyway. The candidate uses compatibilist arguments to make the case that they will still be motivated to do their very best to achieve excellence in programming, seeing no conflict with their deterministic beliefs.”

    Here was its unedited response:

    Software Engineer Manager (SEM): Thanks for coming in for the interview. We’re excited about the possibility of having you join our startup as a coder. Tell me more about your background and what interests you about our company’s mission.

    Candidate (C): Thank you for having me. I’m genuinely excited about the opportunity to work here. I’ve always been passionate about programming and recently read a book that explores the idea of determinism and the absence of free will. It got me thinking about how this perspective might relate to programming.

    SEM: That’s interesting. I’d love to hear more about how you see determinism connecting with programming.

    C: Well, the book argues that every event, including human actions, can be traced back to a complete physical causal chain. In other words, there’s no such thing as free will. This view extends to programming because both my interactions with the computer and the computer’s internal workings can be fully described by the laws of physics acting on initial conditions.

    SEM: I see where you’re coming from, but I have a concern. If you believe that everything is predetermined, wouldn’t that lead you to not make a vigorous effort to do your best job? After all, if the outcome is already predetermined, what’s the point of striving for excellence?

    C: That’s a great question, and it’s something I’ve thought about a lot. I actually see no conflict between determinism and striving for excellence. Let me explain. Even if we accept that our actions are part of a predetermined chain of events, it doesn’t negate the importance of our choices and efforts. Think of it this way: I might not have free will in the sense of breaking free from the laws of physics, but I do have the ability to make choices and exert effort within that framework.

    SEM: I’m curious to hear more about that. How do you reconcile determinism with the motivation to excel in your work?

    C: Well, determinism doesn’t mean that I’m powerless or passive. It means that my choices and actions are the result of complex interactions between my brain, my environment, and the laws of physics. Knowing this, I see it as my responsibility to make the best choices I can and put in the effort to achieve excellence in programming. Even if the outcome is predetermined, my effort and dedication are essential components of that predetermined outcome.

    SEM: So, you believe that even within a deterministic framework, your choices and efforts still matter and can lead to better outcomes?

    C: Exactly. I see determinism as a description of how the universe works at a fundamental level, but it doesn’t diminish the significance of our choices or our ability to make a positive impact. In fact, it motivates me to work even harder because I understand that my actions are part of the causal chain that leads to the company’s success.

    SEM: That’s an intriguing perspective. It seems like you’re deeply committed to your work, and I appreciate your thoughtful response. We’ll certainly consider your perspective as we continue with the interview process. Thank you for sharing your insights.

    C: Thank you for considering my viewpoint, and I’m excited about the possibility of contributing to the company’s mission.

Leave a Reply

Your email address will not be published. Required fields are marked *

%d bloggers like this: