Darth Vader
Article
Darth Vader is a recurring person in the Astral Codex Ten archive, appearing 2 times across 2 issues between November 11, 2022 and January 26, 2023. The archive places it in contexts such as “someone bossed them around particularly badly, and they imagined Darth Vader telling them to give into their anger”; “If you take one step towards Darth Vader, he will take two steps toward you”; “asking GPT to pretend to be … Darth Vader”. It most often appears alongside Jesus, Scott, Aella.
Metadata
- Category: People
- Mention count: 2
- Issue count: 2
- First seen: November 11, 2022
- Last seen: January 26, 2023
Appears In
Related Pages
-
- Jesus (2 shared issues)
-
- Scott (2 shared issues)
-
- Aella (1 shared issues)
-
- aphid (1 shared issues)
-
- astral projection (1 shared issues)
-
- Bayes (1 shared issues)
-
- Bostrom (1 shared issues)
-
- Buddha (1 shared issues)
-
- Buddhists (1 shared issues)
-
- Character.AI (1 shared issues)
-
- ChatGPT (1 shared issues)
-
- ChatGPT (1 shared issues)
External Links
Source Context
Recovered passages from the original issue text. When the raw archive preserved outbound links inside the source passage, they are listed directly under the quote.
Beyond that, the stories are all slightly different, but smoothing them into a single thread: the person got really into some piece of fiction, and found that one of the characters they were modeling really carefully was now “active” in some sense where could give them advice. For example, the person might be kind of a pushover, and then one time after they watched Star Wars ten times in a row, someone bossed them around particularly badly, and they imagined Darth Vader telling them to give into their anger and fight back (this is a fake example, there isn’t someone with a Darth Vader personality running around). Then this became a sufficiently regular occurrence that they could query their “Darth Vader” module and get good advice on social situations that their normal personality wouldn’t give them. Or maybe they would have a Darth Vader monologue running at the same time as their regular monologue, giving different advice. Eventually their inner Darth Vader absorbed various aspects repressed from their regular personality and became a different person from movie-Vader, more fleshed-out and focused on their particular concerns. They emphasize that it really feels like Vader is in their head giving them advice, or that they sometimes “become” Vader - and in particular they emphasize that this is different from just asking themselves “what would Darth Vader do in this situation?”. They understand that most people learning about their situation would expect that they’re exaggerating a much more boring “just ask yourself what Vader would do” situation, and they’re fine with people believing that if they want, but insist that it’s actually something different and more interesting than that.
(I want to repeat that Darth Vader is a fake example - but it does tend to be characters who are very different from their usual personality, and 2/3 of them say the most striking example of this was with an evil character)
In most of the cases I know about, there was some deliberate attempt to cultivate the alternate personality, in the context of other people in the community telling them this might happen and was interesting. A skeptic could seize on this “Aha! I knew it was just people trying to be quirky!” But in the stories these people told me, it was more about - they found that this effort was producing something unexpected, and developing new personality aspects that they needed, so they kept going. If you take one step towards Darth Vader, he will take two steps toward you (sorry if I am sounding like a Sith youth pastor).
But the essay brings up another connotation: to simulate is to pretend to be something. A simulator wears many masks. If you ask GPT to complete a romance novel, it will simulate a romance author and try to write the text the way they would. Character.AI lets you simulate people directly, asking GPT to pretend to be George Washington or Darth Vader.
Inline links: Character.AI
GPT doesn’t really like me. And it’s not lying, saying it likes me when it really doesn’t. It’s simulating a character, deciding on the fly how the character would answer this question, and then answering it. If this were Character.AI and it was simulating Darth Vader, it would answer “No, I will destroy you with the power of the Dark Side!” Darth Vader and the-character-who-likes-me-here are two different masks of GPT-3.
This answer is exactly as fake as the last answer where it said it liked me, or the Darth Vader answer where it says it wants to destroy me with the power of the Dark Side. It’s just simulating a fake character who happens to correspond well to its real identity.