This was written with AI assistance, and my workflow was to give the general theme to AI, have it write an outline, then each chapter, then manually reorganize the text where needed, request major changes, point out subpar sentences/paragraphs for it to rewrite, and do small manual changes. The AI used was mostly Claude 3.5 Sonnet, which seems significantly better than ChatGPT-4o and Gemini 1.5 Pro at this kind of thing.
Thanks! I was unsure whether to post this, because I’m worried about contributing to a future where AI generated content overwhelms everything, and I wasn’t sure the near-term value of the story is high enough to justify posting it anyway. (I thought it was a nice story, but am likely biased. Also asked some AI chatbots, who gave positive reviews, but the risk of them being biased is also very high.)
Hard to copy-paste the whole prompt, because there was a lot of back-and-forth. I maxed out the query limits on several free accounts I created for my family, over several days, to produce the story. Maybe I’ll just describe what I did in more detail. Started with
Can you write an outline for an episode of STTNG in which the crew becomes suspicious of various logical inconsistencies in the universe (including backups being taboo) and starts questioning the nature of their reality?
which resulted in this initial outline:
I. Teaser
The Enterprise responds to a distress call from a Federation outpost
Upon arrival, they find the outpost abandoned, with only a mysterious device left behind
II. Act One
Data activates the device, which emits a pulse of energy
Crew members begin noticing small inconsistencies in their daily routines
Geordi questions why they don’t use replicators to solve resource shortages on colonies
III. Act Two
More crew members report strange observations
Dr. Crusher wonders why they don’t use transporter technology to cure diseases or injuries
Worf ponders the improbability of so many humanoid species evolving independently
Picard becomes concerned about the crew’s unusual behavior
IV. Act Three
Riker raises the question of why they don’t use transporters for backing up crew members
The senior staff meets to discuss the mounting inconsistencies
Data theorizes that the mysterious device has enhanced their critical thinking abilities
Picard orders the ship to return to the outpost for further investigation
V. Act Four
At the outpost, they discover records of experiments in “cognitive enhancement”
The crew starts questioning the nature of their reality
Some crew members experience existential crises
The ship’s computer begins to malfunction as the crew’s doubts intensify
VI. Act Five
Data determines that their enhanced cognition is causing a feedback loop in the ship’s systems
The crew must find a way to deactivate the device and return to normal
Philosophical debate ensues about whether they should retain their newfound awareness
Picard makes the difficult decision to deactivate the device
VII. Epilogue
The crew returns to normal, but with a lingering sense of unease
Picard records a captain’s log reflecting on the experience and the nature of their reality
Final scene shows a subtle glitch in the ship’s operations, leaving a sense of ambiguity
Then the main things I recall doing:
Ask AI to add parts about audience not having critical thinking, and the crew trying to infer other attributes of the audience and writers.
Add second part of conversation between Guinan and Ensign, with both of them having second thoughts about what Guinan initially said.
Add part about suspending ship computer to sandbox, to make clear it wasn’t harmed.
Ask AI for additional realizations/inferences the crew might have, and add selected ones to the story.
Ask AI to change the ending so that the crew retained their new abilities, instead of returning to normal.
Pointing out a bunch of sentences/paragraphs that didn’t make as much sense as they could, with AI generally able to improve them a lot in one try.
Also tried to get AI to find issues with the story on its own, but that didn’t work very well.
Manually make some edits, like changing a word here and there, deleting parts that were repetitive, moving parts arounds to make the story flow better.
So, it wasn’t your idea for the crew to start theorizing that they were fictional characters? Nor was it your idea for them to theorize that they were fictional characters in a story written by AI?
Yeah, it looks like I did not directly tell the AI to have the crew start theorizing that they are fictional characters. After the initial outline (which did not mention fictional characters) was produced, I fed it back to a new instance of Claude and asked:
Please read this outline and suggest some ways to improve the script, especially to make it more philosophically and intellectually interesting.
And as part of its answer it suggested:
Have characters question whether their universe is a simulation, a holodeck program, or even a television show.
Then I wrote
What about exploring their previous lack of critical thinking and implications of that both for their own universe and the larger universe they’re part of?
Which made it focus more on the fiction hypothesis.
The part about the story being written by AI was its own idea. No hints from me as far as I can remember.
ETA: Here’s the whole answer to my “make it more philosophically and intellectually interesting” prompt:
This outline presents an intriguing concept that has the potential to delve deep into philosophical and intellectual territory. Here are some suggestions to enhance the script and make it more philosophically and intellectually engaging:
Expand on the nature of reality:
Introduce the simulation hypothesis early on, perhaps through a casual conversation between Data and Geordi.
Have characters question whether their universe is a simulation, a holodeck program, or even a television show.
Explore the implications of living in a simulated reality and how it affects free will and personal identity.
Deepen the existential crisis:
Have characters grapple with the meaning of their existence if their reality is artificial.
Explore how this realization affects their relationships, duties, and sense of purpose.
Include scenes where characters contemplate whether their memories and experiences are “real.”
Introduce a moral dilemma:
If they discover their reality is simulated, should they attempt to contact their “creators”?
Explore the ethics of continuing to exist in a simulated world vs. attempting to “wake up” to a potentially harsher reality.
Have characters debate the morality of creating simulated universes themselves.
Expand on the scientific implications:
Have Data and Geordi theorize about the computational power required to simulate their universe.
Explore how the laws of physics in their universe might be “programmed” and if they can be altered.
Discuss the anthropic principle and fine-tuning of universal constants.
Incorporate elements of epistemology:
Have characters question how they can know anything for certain given their potentially simulated existence.
Explore concepts like Descartes’ “I think, therefore I am” in the context of artificial consciousness.
Discuss the nature of consciousness and whether it can be truly simulated.
Add a temporal element:
Introduce time dilation or time loops as the crew’s enhanced cognition allows them to perceive time differently.
Explore how their perception of time affects their understanding of causality and free will.
Enhance the ending:
Instead of fully returning to normal, have some characters retain a degree of awareness or doubt.
End with a philosophical discussion between Picard and Data about the nature of reality and the value of ignorance vs. knowledge.
Consider a more ambiguous ending that leaves the question of their reality’s nature unresolved.
Include references to philosophical works:
Have characters discuss relevant philosophical concepts, such as Plato’s Allegory of the Cave, Berkeley’s idealism, or Putnam’s brain in a vat thought experiment.
Include a scene where the crew debates different philosophical perspectives on the nature of reality.
Explore the ethical implications of their technology:
Expand on Riker’s question about using transporters for backing up crew members, discussing the philosophical implications of creating copies of consciousness.
Delve deeper into the ethics of using replicators and how it affects societal structures and values.
Incorporate a meta-narrative element:
Have characters become aware of narrative tropes and question why their lives seem to follow certain storytelling patterns.
Explore the idea of narrative causality and how it might shape their reality.
These suggestions aim to deepen the philosophical and intellectual content of the script while maintaining the core Star Trek elements. The enhanced script would not only entertain but also provoke thought and discussion among viewers about the nature of reality, consciousness, and existence.
I’m sharing a story about the crew of Enterprise from Star Trek TNG[1].
This was written with AI assistance, and my workflow was to give the general theme to AI, have it write an outline, then each chapter, then manually reorganize the text where needed, request major changes, point out subpar sentences/paragraphs for it to rewrite, and do small manual changes. The AI used was mostly Claude 3.5 Sonnet, which seems significantly better than ChatGPT-4o and Gemini 1.5 Pro at this kind of thing.
getting an intelligence/rationality upgrade, which causes them to deduce their fictional nature.
Nice story! Could you copy-paste here the prompt you used? I’m curious what bits of plot were added by you and which were improvised by the AI.
Thanks! I was unsure whether to post this, because I’m worried about contributing to a future where AI generated content overwhelms everything, and I wasn’t sure the near-term value of the story is high enough to justify posting it anyway. (I thought it was a nice story, but am likely biased. Also asked some AI chatbots, who gave positive reviews, but the risk of them being biased is also very high.)
Hard to copy-paste the whole prompt, because there was a lot of back-and-forth. I maxed out the query limits on several free accounts I created for my family, over several days, to produce the story. Maybe I’ll just describe what I did in more detail. Started with
which resulted in this initial outline:
I. Teaser
The Enterprise responds to a distress call from a Federation outpost
Upon arrival, they find the outpost abandoned, with only a mysterious device left behind
II. Act One
Data activates the device, which emits a pulse of energy
Crew members begin noticing small inconsistencies in their daily routines
Geordi questions why they don’t use replicators to solve resource shortages on colonies
III. Act Two
More crew members report strange observations
Dr. Crusher wonders why they don’t use transporter technology to cure diseases or injuries
Worf ponders the improbability of so many humanoid species evolving independently
Picard becomes concerned about the crew’s unusual behavior
IV. Act Three
Riker raises the question of why they don’t use transporters for backing up crew members
The senior staff meets to discuss the mounting inconsistencies
Data theorizes that the mysterious device has enhanced their critical thinking abilities
Picard orders the ship to return to the outpost for further investigation
V. Act Four
At the outpost, they discover records of experiments in “cognitive enhancement”
The crew starts questioning the nature of their reality
Some crew members experience existential crises
The ship’s computer begins to malfunction as the crew’s doubts intensify
VI. Act Five
Data determines that their enhanced cognition is causing a feedback loop in the ship’s systems
The crew must find a way to deactivate the device and return to normal
Philosophical debate ensues about whether they should retain their newfound awareness
Picard makes the difficult decision to deactivate the device
VII. Epilogue
The crew returns to normal, but with a lingering sense of unease
Picard records a captain’s log reflecting on the experience and the nature of their reality
Final scene shows a subtle glitch in the ship’s operations, leaving a sense of ambiguity
Then the main things I recall doing:
Ask AI to add parts about audience not having critical thinking, and the crew trying to infer other attributes of the audience and writers.
Add second part of conversation between Guinan and Ensign, with both of them having second thoughts about what Guinan initially said.
Add part about suspending ship computer to sandbox, to make clear it wasn’t harmed.
Ask AI for additional realizations/inferences the crew might have, and add selected ones to the story.
Ask AI to change the ending so that the crew retained their new abilities, instead of returning to normal.
Pointing out a bunch of sentences/paragraphs that didn’t make as much sense as they could, with AI generally able to improve them a lot in one try.
Also tried to get AI to find issues with the story on its own, but that didn’t work very well.
Manually make some edits, like changing a word here and there, deleting parts that were repetitive, moving parts arounds to make the story flow better.
So, it wasn’t your idea for the crew to start theorizing that they were fictional characters? Nor was it your idea for them to theorize that they were fictional characters in a story written by AI?
Yeah, it looks like I did not directly tell the AI to have the crew start theorizing that they are fictional characters. After the initial outline (which did not mention fictional characters) was produced, I fed it back to a new instance of Claude and asked:
And as part of its answer it suggested:
Then I wrote
Which made it focus more on the fiction hypothesis.
The part about the story being written by AI was its own idea. No hints from me as far as I can remember.
ETA: Here’s the whole answer to my “make it more philosophically and intellectually interesting” prompt:
This outline presents an intriguing concept that has the potential to delve deep into philosophical and intellectual territory. Here are some suggestions to enhance the script and make it more philosophically and intellectually engaging:
Expand on the nature of reality:
Introduce the simulation hypothesis early on, perhaps through a casual conversation between Data and Geordi.
Have characters question whether their universe is a simulation, a holodeck program, or even a television show.
Explore the implications of living in a simulated reality and how it affects free will and personal identity.
Deepen the existential crisis:
Have characters grapple with the meaning of their existence if their reality is artificial.
Explore how this realization affects their relationships, duties, and sense of purpose.
Include scenes where characters contemplate whether their memories and experiences are “real.”
Introduce a moral dilemma:
If they discover their reality is simulated, should they attempt to contact their “creators”?
Explore the ethics of continuing to exist in a simulated world vs. attempting to “wake up” to a potentially harsher reality.
Have characters debate the morality of creating simulated universes themselves.
Expand on the scientific implications:
Have Data and Geordi theorize about the computational power required to simulate their universe.
Explore how the laws of physics in their universe might be “programmed” and if they can be altered.
Discuss the anthropic principle and fine-tuning of universal constants.
Incorporate elements of epistemology:
Have characters question how they can know anything for certain given their potentially simulated existence.
Explore concepts like Descartes’ “I think, therefore I am” in the context of artificial consciousness.
Discuss the nature of consciousness and whether it can be truly simulated.
Add a temporal element:
Introduce time dilation or time loops as the crew’s enhanced cognition allows them to perceive time differently.
Explore how their perception of time affects their understanding of causality and free will.
Enhance the ending:
Instead of fully returning to normal, have some characters retain a degree of awareness or doubt.
End with a philosophical discussion between Picard and Data about the nature of reality and the value of ignorance vs. knowledge.
Consider a more ambiguous ending that leaves the question of their reality’s nature unresolved.
Include references to philosophical works:
Have characters discuss relevant philosophical concepts, such as Plato’s Allegory of the Cave, Berkeley’s idealism, or Putnam’s brain in a vat thought experiment.
Include a scene where the crew debates different philosophical perspectives on the nature of reality.
Explore the ethical implications of their technology:
Expand on Riker’s question about using transporters for backing up crew members, discussing the philosophical implications of creating copies of consciousness.
Delve deeper into the ethics of using replicators and how it affects societal structures and values.
Incorporate a meta-narrative element:
Have characters become aware of narrative tropes and question why their lives seem to follow certain storytelling patterns.
Explore the idea of narrative causality and how it might shape their reality.
These suggestions aim to deepen the philosophical and intellectual content of the script while maintaining the core Star Trek elements. The enhanced script would not only entertain but also provoke thought and discussion among viewers about the nature of reality, consciousness, and existence.