So, I was walking through the hallway this morning and I nearly tripped over a stack of stanchions. You know, those heavy velvet rope barriers you see at movie theaters or fancy clubs? It turns out our housemate Daniel picked them up from a clothing store that was closing down. He said they were a bargain, but now they are just sitting there, taking up space in the apartment. It got me thinking about how we hold onto things, or ideas, or even entire career paths, just because we have already put something into them.
Herman Poppleberry at your service. And Corn, you are hitting on one of my absolute favorite topics in behavioral economics and psychology. Daniel actually sent us a prompt about this very thing. He was talking about how he finally threw out a massive stack of old optical media, like compact discs and digital versatile discs, that he had been hauling around for years. He even mentioned how he spent so much effort importing a specialized burner to Israel a few years back, and that effort made it almost impossible for him to let go of the discs, even though he has not used them in forever.
It is the classic sunk cost fallacy. We feel like if we stop now, all that previous effort or money was wasted. But the reality is, that money is gone regardless of what we do next. So, why is it so hard for our brains to accept that?
It is fascinating because it is deeply counter-intuitive from a purely logical standpoint. If you are at a movie and you realize twenty minutes in that it is absolutely terrible, the logical thing to do is leave. You have already spent the fifteen dollars for the ticket. That money is gone. If you stay for the remaining ninety minutes, you are not getting your money back, you are just losing ninety minutes of your life on top of the fifteen dollars. But most people will sit through the whole thing because they feel they need to get their money's worth.
I see this in professional settings all the time, too. Daniel mentioned it in the context of artificial intelligence development. He has been working with these coding agents lately, and he noticed they get into these tunnel vision loops. An agent will try to fix a bug, fail, and then instead of stepping back and saying, wait, maybe this whole approach is wrong, it just keeps doubling down on the same broken logic, trying to patch the patch.
That is such a great observation. It is almost like the artificial intelligence is inheriting our human stubbornness. In software engineering, we call it the "throwing good money after bad" problem. But with these agentic systems in twenty twenty-six, it is more like "throwing good inference compute after bad." We actually touched on the way these agents communicate back in episode two hundred eighteen, when we talked about the agentic mesh. The problem is that once a system has a plan, the cost of re-planning often feels higher than the cost of just trying one more time to make the current plan work. We are seeing this even in the latest reasoning models; they get stuck in a recursive loop of their own "Chain of Thought" and can't find the exit door.
Is there a biological reason for this? I mean, we are brothers, we have both seen each other get stubborn about things. Is there something in the way our brains are wired that makes us want to double down rather than pivot?
There absolutely is. It is tied to something called loss aversion. Evolutionarily speaking, for our ancestors, losing a resource was often much more dangerous than the potential benefit of gaining a new one. Losing your winter food supply could mean death, whereas finding a little extra food was just a bonus. So, our brains are hardwired to feel the pain of a loss about twice as intensely as the joy of an equivalent gain. When we abandon a project we have invested in, our brain registers that as a loss. It hurts. Doubling down feels like a way to avoid that pain, even if it is statistically likely to lead to a bigger loss later.
So we are essentially gambling with our time because we are afraid to admit the initial bet failed. That makes sense. But Daniel's prompt was asking for hacks. How do we actually break out of this? Especially when it is something big, like a career. We have talked about the architecture of resilience in episode three hundred fourteen, but how do you apply that to your own life when you realize the path you are on is no longer fit for purpose?
One of the most effective psychological hacks is something called the "Time-Traveler Test" or the "Clean Slate" approach. You ask yourself: if I walked into this situation today for the first time, with no prior history, would I choose to invest in this? If you were looking at your career today, starting from scratch, would you pick the same one? If the answer is no, then the only reason you are still doing it is the sunk cost.
That is a tough one to face. It requires a lot of intellectual honesty. I think another version of that is the "Adviser Hack." We are usually much better at giving advice to others than to ourselves. If a friend came to you and said, I have spent five years on this project, it is not working, and I am miserable, what would you tell them? You would almost certainly tell them to quit. But when it is us, we get blinded by our own internal narrative.
Exactly. In psychology, that is known as Solomon's Paradox. We reason more wisely about other people's problems because we are not emotionally entangled in their sunk costs. We can see the logic clearly when it is not our own skin in the game. But here is where it gets really interesting in the world of artificial intelligence. We are actually starting to build "supervisory layers" into these systems. In modern twenty twenty-six development, we have meta-agents whose only job is to monitor the primary agent and say, hey, you have been trying the same thing for ten iterations and the error rate is not dropping. Stop. Re-evaluate.
We need personal meta-agents. Maybe that is what friends are for. But let's go deeper into the "doubling down" aspect. There is a fine line between the sunk cost fallacy and actual persistence. We praise people for being "gritty" and sticking with things when they get hard. How do you distinguish between being a hero who persevered and being the person who stayed in the theater for a movie they hated?
That is the million-dollar question, Corn. The difference usually lies in the feedback loop. Persistence is sticking with a goal while being flexible about the methods. The sunk cost fallacy is sticking with a specific method even when the goal is no longer achievable or the method is clearly failing. If you are a marathon runner and your legs are tired, you persist. If you are a marathon runner and you have a compound fracture in your leg, you are not persisting by continuing to run; you are succumbing to a fallacy that will cause permanent damage.
I like that distinction. It is about the data. If the data says the approach is fundamentally broken, then continuing is irrational. Think about the Concorde supersonic jet. That is the most famous historical example. The British and French governments knew the plane was never going to be economically viable quite early on, but they kept pouring billions into it because they had already spent so much and didn't want to lose face. It became known as the "Concorde Fallacy."
And face-saving is a huge part of this. It is not just about the money or the time; it is about our ego. We don't want to look like we failed. But in the tech world, especially in the last year or two, we have seen a shift toward the "fail fast" mentality. It is better to spend ten thousand dollars and realize a product is a dud than to spend ten million dollars trying to convince people to buy something they don't want.
Let's talk about physical objects for a second, because that is where Daniel started with his optical media and those stanchions in our hallway. Why do we feel such a strong connection to physical stuff we have "invested" in? He mentioned the effort of importing that burner to Israel. Is it the effort itself that creates the value?
Yes, that is actually known as the "IKEA Effect." Research shows that people value products more highly if they have put effort into creating or obtaining them. It is rooted in our sense of self-efficacy; we feel capable because we did the work. Daniel didn't just buy a burner; he navigated the complexities of international shipping and Israeli customs. That "labor" is now baked into his perception of the object's value. It makes the object feel like a part of his identity.
So, when he throws out the discs, he feels like he is throwing out a piece of his own history and effort. That is a heavy psychological burden for a piece of plastic.
It really is. A great hack for physical clutter is the "Reverse Packing Party." You act as if you are moving. You put everything in boxes. If you don't take it out of the box for three months, you don't actually need it. The fact that you paid for it five years ago is irrelevant to your life today. It is about the carrying cost of the space it occupies.
Speaking of things people pay for, remember episode three hundred eleven where we talked about that five thousand dollar defamation case over a pizza review? That felt like a sunk cost situation too. Once the legal fees started piling up, both sides probably felt they had to keep going to "win," even though the original dispute was tiny.
Legal battles are the ultimate sunk cost traps. You spend ten thousand dollars on a lawyer to win a five thousand dollar settlement. It makes zero sense, but people do it all the time because they want "justice" or they want to be "right." But "right" is often the most expensive thing you can be.
So, how do we apply this to careers? That feels like the highest stakes area. You spend four years in university, maybe another three in a specialized graduate program, and then five years in the field. You are twelve years in. You realize you are miserable. The thought of starting over at thirty-five or forty feels impossible because you "wasted" those twelve years.
But you didn't waste them. You gained skills, perspective, and maturity. The only way to actually waste the next twenty years is to stay in a career you hate just because of the twelve years you already spent. One of my favorite ways to frame this is "Zero-Based Career Budgeting." Every year, you should look at your career and ask: if I were not in this job today, would I apply for it? If the answer is no, you need an exit strategy. It doesn't mean you quit tomorrow, but it means you stop investing your identity into it.
I think people also worry about what others will think. "Oh, did you hear about Herman? He was a successful donkey engineer and now he is starting a bakery? What a waste of his degree!" We are so afraid of that judgment.
People are going to judge you regardless. But most people are too busy worrying about their own sunk costs to actually care about yours for more than a few minutes. Another hack is to reframe the "loss" as a "tuition payment." You didn't lose five years on a failing startup; you paid five years of tuition to learn how not to run a company. That shift from "loss" to "investment in knowledge" can take the sting out of moving on.
That is a great way to put it. It turns a negative narrative into a positive one. Let's look at the AI side of this again. As we move further into twenty twenty-six, we are seeing more autonomous agents doing long-term tasks. If we don't solve the sunk cost problem for AI, we are going to have systems that waste incredible amounts of energy and money pursuing dead ends.
We are already seeing solutions for this. There is a concept called "Recursive Oversight." You have one model doing the work, a second model reviewing the work, and a third model looking at the "trajectory" of the first two. If the trajectory is flat, meaning no progress is being made, the third model has the authority to kill the process. We need to do that for ourselves. We need to look at the trajectory of our happiness, our productivity, or our bank accounts. If the line has been flat or declining for a long time, the "process" needs to be killed.
It sounds so cold when you put it that way, "killing the process," but it is actually an act of self-mercy. It is giving yourself permission to try something else. I think about Daniel and his stanchions. If he keeps them, he is "paying" for them every day with the frustration of tripping over them. If he gives them away, he stops paying.
Exactly. Everything has a "carrying cost." Relationships, old hobbies, clothes that don't fit, even old grudges. They all take up mental or physical bandwidth. When you realize that the cost of keeping something is higher than the benefit it provides, the logical choice is to let go.
Let's talk about the psychological "hack" of pre-commitment. I have heard that setting a "quit date" or a "stop-loss" point before you start something can help. For example, if I am starting a new business, I might say, "If I am not profitable in eighteen months, I will close it down." Does that actually work, or do we just move the goalposts when the time comes?
We definitely try to move the goalposts. That is the human way! But a pre-commitment is much harder to ignore if you make it public or if you involve a neutral third party. This is why "Mastermind" groups or even just a good friend can be so helpful. You tell them, "Hey, if I am still complaining about this job in six months, I want you to sit me down and tell me it is time to leave." You give them permission to be your "circuit breaker."
I like that. It is like an externalized conscience. But what about the role of "sunk cost" in relationships? That is probably the most painful one. People stay together for years because they have "put so much time into it," even when they are clearly no longer compatible.
That is perhaps the most tragic application of the fallacy. People often say, "I don't want to have wasted the last ten years." But by staying another ten years in a broken relationship, they are just doubling the "waste." They are valuing the past more than their future. It is a form of temporal myopia. We get so focused on the years behind us that we can't see the decades in front of us.
It feels like the ultimate hack is a shift in perspective on time. We think of time as something we have "spent," like money in a bank account. But time is actually more like a river. It is always flowing. You can't get the water back that has already passed the bridge, but you can decide which direction you are going to swim in the water that is coming toward you.
That is very poetic, Corn. And scientifically, it is more accurate. The past is fixed. The "cost" is truly sunk. It is at the bottom of the ocean. The only thing you can optimize is the future. Every morning you wake up, you are essentially given a fresh set of resources: your time, your energy, and your attention. The smartest thing you can do is allocate those resources where they will have the highest return from this moment forward, regardless of what you did yesterday.
So, for our listeners who are feeling stuck, whether it is a project at work, a cluttered garage, or a career that feels like a dead end, what is the first step?
The first step is a "Sunk Cost Audit." Make a list of the things in your life that you are doing primarily because you have "always done them" or because you have "invested so much." Then, apply the Time-Traveler Test to each one. If you started today with a clean slate, would I choose this? If the answer is no, then you have identified a sunk cost trap.
And then, you have to be brave enough to step out of the trap. It is going to hurt. Your brain is going to scream that you are losing something. You have to acknowledge that pain, but don't let it drive the car. Let the logic and the "future-you" drive the car.
And remember that "giving up" on a bad path is actually "starting" on a better one. It is not a failure; it is a pivot. In the world of twenty twenty-six, where everything moves so fast, the ability to pivot is actually one of the most valuable skills you can have. Whether you are a human or a Large Language Model, the capacity to realize you are on the wrong track and change course is the hallmark of true intelligence.
I think that is a perfect place to wrap this up. Daniel, if you are listening, please do something about those stanchions. I love you, man, but my shins can't take it anymore. And to our listeners, thank you for sticking with us through episode three hundred sixteen.
If you found this discussion helpful, or if you have a story about a time you successfully beat the sunk cost fallacy, we would love to hear from you. You can find us at myweirdprompts.com where there is a contact form, and of course, you can subscribe to our feed there as well.
And hey, if you have been enjoying the show, a quick review on your podcast app or on Spotify really helps us out. It helps other curious people find these deep dives and helps us keep the lights on here in Jerusalem.
It really does make a difference. We appreciate every one of you who tunes in. This has been My Weird Prompts.
Until next time, keep asking those weird questions and don't be afraid to leave the theater if the movie is bad.
See you next week!
