I was scrolling through some old reviews last night and it hit me how much the landscape of failure has changed recently. Today is March sixteenth, twenty twenty-six, and looking back at the last three years of cinema, it feels like we have entered a strange new era. Today's prompt from Daniel is about the worst movies from the last three years, specifically the ones that failed unintentionally. We are talking about the massive, high-budget swings that did not just miss the mark, but somehow managed to redefine what a disaster looks like in the age of the algorithm.
Herman Poppleberry here, and I have been waiting for this. We touched on the sincerity threshold back in episode eleven seventy-one, the idea that for a movie to be a truly great disaster, it has to really try. It cannot be a wink-and-a-nod spoof. It cannot be self-aware. It has to be a hundred-million-dollar swing that misses the ball so hard the bat flies into the stands and takes out the mascot. We are looking for that specific intersection of high-budget hubris and creative blindness.
It is a specific kind of hubris, right? We are in this era where AI can generate technically perfect images, yet human-led studios are still putting out these massive, earnest catastrophes. It feels like a golden age for people who love to watch a train wreck. There is something almost refreshing about a human being making a mistake this expensive. It is a reminder that even with all the data in the world, you cannot manufacture soul, and you certainly cannot hide the lack of one with more CGI.
It really is. To qualify for our list today, we are looking at releases from twenty twenty-three through early twenty twenty-six. They need to have those abysmal Rotten Tomatoes scores, usually sub-fifteen percent, though we might make an exception for a few that were so culturally disastrous they transcend a mere percentage. Most importantly, they have to be devoid of self-awareness. No Sharknado, no Scary Movie, no Winnie the Pooh: Blood and Honey. We want the stuff where the director thought they were making a masterpiece, where the actors thought they were winning Oscars, and the studio thought they were printing money.
Why do we enjoy this so much? I mean, I know why I do. There is something comforting about seeing a massive corporation spend two hundred million dollars on something that makes less sense than a toddler's finger painting. It levels the playing field. It says that money and resources do not guarantee quality.
There is a psychological release in it. We live in a world of highly polished, algorithmically optimized content. Everything is focus-grouped to death. When something fails this spectacularly, it feels human. It breaks the simulation. It reminds us that despite all the data and the focus groups, nobody actually knows anything. It is the ultimate proof that art is not a science.
Let's get into the criteria then. We are looking for narrative voids masked by CGI, miscast stars, and that specific scent of development hell. You can almost smell the studio interference on some of these. It is like a layer of film over the lens that just screams "too many cooks in the kitchen."
The smell of desperation and reshoots. Let's start the countdown at number ten with The Flash from twenty twenty-three. Now, critics were actually kinder to this than they should have been, but it represents the absolute peak of the CGI-over-script failure. This was in development for nearly a decade. Multiple directors, multiple scripts, and a lead actor who was... let's say, problematic in the press. By the time it came out, the CGI looked like something from a nineteen-nineties video game.
That was the weirdest part. They claimed the distorted faces in the speed force were an intentional artistic choice. I remember thinking, do they really think we are that gullible? It looked like the characters were melting. It was a visual nightmare that cost over two hundred million dollars. When you spend ten years on a project and the final product looks like a beta version of a PlayStation two game, you have failed the sincerity threshold.
That is the quintessential example of trying to fix a movie in post-production. They spent so much time trying to salvage the footage they had that they forgot to give the characters a reason to exist. It had a fifteen percent drop in its second weekend and just never recovered. It is the poster child for a movie that was too expensive to fail, yet too incompetent to succeed. It was the end of an entire cinematic universe, and it went out with a whimper and a very poorly rendered cameo of Nicolas Cage.
Moving to number nine, we have Expendables four, or Expend-four-bles as they tried to stylize it. This one felt like a direct-to-video movie that somehow escaped into theaters with a hundred-million-dollar price tag. It currently holds a fourteen percent on Rotten Tomatoes, and honestly, that feels generous.
The green screen work in that movie was offensive. There are scenes where characters are supposedly on a massive ship, but you can tell they are standing in a small room in Bulgaria with a fan blowing on them. The lighting does not match, the shadows are all wrong, and the action is edited so frantically you cannot tell who is hitting whom. It lacked the charm of the earlier entries and just felt tired. It is the "Star-Power Vacuum" in action. You have all these big names, or names that used to be big, and they are just going through the motions for a paycheck.
It is that vacuum we talked about. There is no chemistry, just a series of explosions that do not feel like they have any weight. It is a movie made of tropes that were already old in nineteen eighty-five. When you see Sylvester Stallone and Jason Statham looking bored in their own action movie, the audience is going to be bored too.
Number eight is Argylle from early twenty twenty-four. Apple spent two hundred million dollars on this. It was supposed to be the start of a massive franchise, a new kind of spy thriller. Instead, we got a convoluted mess with a CGI cat and Henry Cavill in a flat-top haircut that haunted my dreams. It sits at a thirty-three percent, but in terms of "unintentional disaster," it is a heavy hitter because of the sheer scale of the miscalculation.
The twists! Every ten minutes there was a new twist that made the previous ten minutes irrelevant. By the end, I did not care who was a spy or who was a novelist. I just wanted it to be over. It is a perfect example of a director having too much budget and no one to tell him no. Matthew Vaughn is a talented guy, but Argylle felt like he was playing with a two-hundred-million-dollar chemistry set and accidentally blew up the lab.
That is the danger of the streaming wars. These tech companies like Apple and Netflix have so much cash that they throw it at big-name directors without any oversight. They want "content" for the platform, but they forget that movies need a cohesive vision. The result is a movie that feels like a fever dream, where the tone shifts from slapstick comedy to gritty spy action every five seconds.
Let's talk about number seven, which is a two-for-one special. Rebel Moon Part One and Part Two. Zack Snyder's space epic that was originally a Star Wars pitch. It is essentially five hours of slow-motion wheat harvesting and exposition dumps.
The sincerity there is off the charts. Snyder truly believes he is making the next Mythology with a capital M. But the characters are cardboard. There is a scene where a character talks to a robot for ten minutes about the nature of joy, and it feels like reading a philosophy textbook written by a teenager who just discovered Nietzsche. It is the "Development Hell" cycle where a creator spends so long in their own head that they forget how humans actually talk.
And yet, it has this incredible visual fidelity in some spots, and then looks like a muddy mess in others. It is the ultimate "sincerity threshold" movie because it is so earnest and yet so profoundly boring. It is a movie that demands you take it seriously while giving you absolutely no reason to do so.
Number six is the one everyone expected. Madame Web. Released in twenty twenty-four, holding a steady eleven percent on Rotten Tomatoes. This movie is a miracle of incompetence. The dialogue sounded like it was translated into three different languages and then back into English by a broken LLM.
"He was in the Amazon with my mom when she was researching spiders right before she died." That line will live in infamy. What fascinates me about Madame Web is that you can see the moments where they clearly changed the plot in post-production. The actors' mouths do not match the words they are saying in half the scenes. It is like watching a dubbed foreign film, except it was filmed in English.
That is the development hell cycle in real-time. They realized the original plot, which supposedly involved a nineteen-nineties period piece and a different villain, did not work. So they tried to rewrite the movie using ADR, which is automated dialogue replacement. It creates this uncanny, disjointed feeling where you feel like you are watching a movie that is actively fighting against its own existence. It is the corporate committee movie. You can see the fingerprints of forty different executives all trying to steer the ship in different directions while the ship is already underwater.
This leads us perfectly into the top five. Number five is Borderlands from late twenty twenty-four. This was a massive intellectual property, a huge fan base, and they cast Cate Blanchett and Kevin Hart. On paper, maybe it works? In reality, it was a disaster. It holds a ten percent on Rotten Tomatoes and lost over seventy percent of its box office revenue in the second weekend.
It felt like a movie made by people who had seen a picture of the game but had never actually played it. It was loud, it was colorful, and it was completely hollow. It tried so hard to be Guardians of the Galaxy but forgot that those movies have heart. This is the "Algorithm-First" approach. The data said people like space heists, they like Kevin Hart, and they like established IP. So they smashed them together regardless of whether it made sense for the world of Borderlands.
I wonder if these movies end up as bait on those illegal streaming sites we discussed in episode eleven seventy-two. Like, nobody wants to pay for Borderlands, but they might click a "watch for free" link and end up with a virus.
I would bet on it. These high-profile flops are perfect malware bait because there is still a high level of curiosity but low willingness to pay. People want to see how bad it is, but they do not want to give the studio twenty dollars for the privilege. That curiosity is exactly what the "Pirate's Trap" exploits.
Number four on the list is Joker: Folie à Deux. This one is controversial because the first movie was such a massive hit. But the sequel was a jukebox musical that seemed to actively dislike its own audience. It cost two hundred million dollars and made a fraction of that, sitting at a thirty-two percent.
It was a fascinating failure. The choice to turn it into a musical where the musical numbers are all imaginary was a bold creative swing that landed flat on its face. It felt like the director was trying to deconstruct the character in a way that left nothing for the fans to hold onto. It is the "Star-Power Vacuum" again, where you have Lady Gaga and Joaquin Phoenix, two of the most talented people on earth, and you give them nothing to do but sing covers of old standards in a prison cell.
It was a very expensive way to tell the fans of the first movie that they were wrong for liking it. That kind of intentional subversion often leads to an unintentional disaster because you alienate your base without gaining a new one. The Rotten Tomatoes score plummeted as soon as people realized what it actually was. It is a movie that thinks it is smarter than the person watching it, which is the quickest way to lose an audience.
Number three is Megalopolis. Francis Ford Coppola spent one hundred and twenty million dollars of his own money on this. It is a movie about a visionary architect in a futuristic New York that looks like ancient Rome. It is completely, utterly insane. It sits at a forty-six percent, but in terms of "unintentional disaster," it is a masterpiece.
There is a scene where a live actor in the theater is supposed to talk to the screen. How do you even do that in a suburban multiplex? It was a movie that refused to follow any rules of narrative or logic. It is the "sincerity threshold" at its peak. He spent decades thinking about this, and the result is something that feels like it was edited by a blender.
I respect Coppola for the swing, but as a movie, it is a catastrophe. It is beautiful in moments, but it is also incomprehensible. It is the ultimate passion project gone wrong. When you have that much legacy and that much money, no one can tell you that the scene with the golden bow and arrow does not make sense. It is a reminder that even the greatest directors need an editor who is not afraid to say "no."
Number two is the twenty twenty-five Snow White remake. This movie was plagued by controversy from the start, but the actual film was a technical nightmare. The CGI dwarves looked terrifying. They tried to bridge the gap between realism and the cartoon aesthetic and ended up in the deepest part of the uncanny valley. It cost over two hundred and fifty million dollars and is currently sitting at a twelve percent.
It felt like a movie that was trying to please everyone and ended up pleasing no one. They changed the story to be more "modern," but in doing so, they lost the simple archetypal power of the original fairy tale. The production budget ballooned because of the constant reshoots to fix the CGI. It is a prime example of how modern editing workflows can actually hurt a film. They keep tweaking things until the soul of the movie is completely gone, replaced by a gray, digital slurry. It is the "corporate committee" failure at its most expensive.
And finally, our number one. The Minecraft Movie from twenty twenty-five. This was the ultimate "what were they thinking" moment. Taking a game made of blocks and putting Jack Black in a blue shirt in a hyper-realistic world was a choice that will be studied in film schools for years as a warning. It holds a nine percent on Rotten Tomatoes.
It looked like a high-budget commercial for a product that does not exist. The visual style was jarring. It was not stylized enough to be artistic, and it was not realistic enough to be immersive. It just looked ugly. And for a brand as big as Minecraft, that is an incredible failure. It holds the record for the most-disliked trailer in history for a reason. It was an unintentional disaster that started before the movie even hit theaters.
The kids hated it because it did not look like the game, and the adults hated it because it was a loud, obnoxious mess. It is the perfect storm of a studio not understanding why people like the thing they are adapting. They thought the name "Minecraft" was enough to carry a two-hundred-million-dollar budget. It is the ultimate "Algorithm-First" failure.
Looking at this list, there is a common thread. Most of these movies were made using an "algorithm-first" approach. They look at data points—star power, intellectual property, trending genres—and try to build a movie like a Lego set. But they forget that cinema requires a cohesive vision. The data can tell you what people liked in the past, but it cannot tell you how to make something new that people will love.
I also think the economics of the "flop" have changed. We are seeing more of these movies because the middle-budget film has disappeared. It is either a five-million-dollar indie or a two-hundred-million-dollar blockbuster. When a two-hundred-million-dollar movie misses, it doesn't just miss—it craters. Some studios are even leaning into the tax write-off model. We saw it with Batgirl and Coyote versus Acme. They would rather take the guaranteed money from a tax break than risk a theatrical release that might embarrass them.
That is the part that fascinates me. Someone watched Minecraft or Snow White and said, "Yes, this is ready. This will make us a billion dollars." It is a bubble of yes-men. When you are that deep into a project, you lose all perspective. You have been looking at the same distorted CGI faces for three years, and eventually, they start to look normal to you.
So, for our listeners who want to experience these disasters for themselves, how do they find the "good" bad movies versus the ones that are just a chore to watch?
You have to look for the sincerity. A movie like Megalopolis is a "good" bad movie because it is trying so hard to be profound. It is fascinating to watch Coppola's mind at work, even if it is a mess. A movie like Expendables four is just a chore because it is lazy. There is no effort, so there is no joy in the failure. Use the public data. Look at the gap between the critic score and the audience score on Rotten Tomatoes. If the critics hate it but the audience score is also low, you might have a disaster on your hands. But if the audience score is zero, it is probably just boring. You want that sweet spot of high-budget hubris.
Also, watch the release dates. If a movie has been pushed back three or four times, that is a massive red flag. It means they have been in the edit suite trying to find a movie that isn't there. The Flash and Snow White are perfect examples of this. Shifting release dates usually mean the studio is panicked.
There is also a value in human error. In an increasingly automated world, seeing a human-made disaster is a reminder of our limitations. It is messy, it is weird, and it is unpredictable. Ten years from now, people will still be talking about the "Amazon with my mom" line from Madame Web. No one will remember the generic superhero movie that came out the month after. It is the "Sincerity Threshold" again. You have to care enough to fail that badly.
As we look toward the next few years, I wonder if generative video tools will make these kinds of disasters a thing of the past. If a studio can just prompt an AI to "make the CGI look good" or "fix the pacing," will we still get these weird, idiosyncratic failures?
I suspect we will just get a different kind of failure. An AI might produce something technically perfect but even more soulless. The human element, with all its flaws and ego, is what makes a disaster truly "great." If you take away the ego, you just have a boring, optimized product. The hubris of a director like Coppola or the corporate confusion of a studio like Sony is what creates the "magic." You cannot simulate that kind of chaos.
We have covered a lot of ground today. From the uncanny valley of Snow White to the blocky disaster of Minecraft. It is a reminder that no matter how much money you have, you cannot buy a good story.
And you cannot buy taste.
If you are looking to dive deeper into why we love these flops, definitely go back and listen to episode eleven seventy-one. We really break down the psychology of the viewer there. And if you are interested in the darker side of how these movies are distributed online, episode eleven seventy-two on the Pirate's Trap is a must-listen.
Thanks as always to our producer Hilbert Flumingtop for keeping the show running smoothly. And a big thanks to Modal for providing the GPU credits that power our research and generation.
This has been My Weird Prompts. If you are enjoying the show, a quick review on your podcast app really helps us reach new listeners who might share our love for a good disaster. You can find our full archive and the top ten list with links at myweirdprompts dot com.
See you next time.
Goodbye.
