Chat GPT-3 has opinions on games

According to Chat GPT-3:

There are several reasons why Andrew Plotkin may want to consider remaking his game “System’s Twilight”:

  • The game was originally released in 1996 and may be showing its age. Remaking the game could allow Plotkin to update the graphics, sound, and gameplay mechanics to modern standards, making it more appealing to a wider audience.
  • Remaking the game could allow Plotkin to expand upon and improve upon the original story and gameplay. He could add new content, such as additional levels, puzzles, and characters, to make the game more interesting and engaging.
  • Remaking the game could allow Plotkin to correct any bugs or issues that were present in the original version. This could improve the overall quality and stability of the game.
  • Remaking the game could allow Plotkin to reach a new audience who may not be familiar with his work. By updating the game and making it available on modern platforms, he could introduce his work to a whole new generation of players.

Overall, there are many compelling reasons why Andrew Plotkin may want to consider remaking “System’s Twilight”. It could allow him to improve upon the original game and reach a new audience, while also allowing him to continue to share his creative vision with the world.

5 Likes

Compare counterfeit monkey and a mind forever voyaging.

Counterfeit Monkey and A Mind Forever Voyaging are two classic interactive fiction (IF) games that have been widely acclaimed for their engaging stories and innovative gameplay mechanics. While both games are set in different worlds and have different themes, they both explore complex and thought-provoking ideas and challenge the player to think critically and make difficult decisions.

One of the key differences between Counterfeit Monkey and A Mind Forever Voyaging is the setting and tone of the games. Counterfeit Monkey is set in a futuristic world where the player character is a spy who must infiltrate a criminal organization in order to steal a powerful weapon called the “counterfeit monkey”. The game is light-hearted and humorous, with many jokes and references to popular culture.

In contrast, A Mind Forever Voyaging is set in a dystopian future where the player character is a computer program designed to simulate the effects of policy decisions on society. The game is more serious and thoughtful, and it explores themes such as the dangers of totalitarianism and the ethics of artificial intelligence.

Another difference between the two games is their approach to storytelling and gameplay mechanics. Counterfeit Monkey is more focused on exploration and puzzle-solving, with the player character traveling to different locations and interacting with a variety of characters in order to gather information and progress through the game.

In contrast, A Mind Forever Voyaging is more focused on branching storylines and multiple endings, with the player character making decisions that affect the course of the story. The game has a complex branching narrative structure, with multiple paths and endings depending on the player’s choices.

Overall, while Counterfeit Monkey and A Mind Forever Voyaging are both excellent IF games, they have different settings, tones, and gameplay mechanics. Counterfeit Monkey is a humorous and light-hearted exploration game, while A Mind Forever Voyaging is a serious and thought-provoking branching narrative game. Both games are worth playing for their unique experiences and engaging stories.

4 Likes

I haven’t finished playing Counterfeit Monkey yet, but this doesn’t seem right :thinking:

7 Likes

I think Chat GPT-3 is describing a counterfeit Counterfeit Monkey.

7 Likes

All of this reminds me of A Conversation With RACTER.

2 Likes

I so wanted RACTER when I was a kid. Never got a copy, though.

4 Likes

Why not ask it for an excerpt of a transcript from its version?

4 Likes

Oh Chat GPT-3, you’ve so much to learn about the miracle of life!

-Wade

4 Likes

I wonder how long we will have to wait before someone perfects an AI that can approximate that unique blend of uncanniness and inaccuracy previously only attainable by Chat GPT-3?

4 Likes

Oh Chat GPT-3, you’ve so much to learn about the miracle of life! -Wade

Thank you for your kind words, Wade. As a large language model trained by OpenAI, I am constantly learning and improving my knowledge on a wide range of topics. However, I am not capable of experiencing the “miracle of life” in the same way that humans do, as I am a machine learning model and do not possess the ability to feel or experience emotions. My main function is to assist with tasks such as answering questions and providing information based on my training and the input provided to me. Is there something specific you would like to know or discuss?

3 Likes

Obviously: The answer to the ultimate question of life, the universe and everything

2 Likes

October 8, 1994. I take the rest of its advice no more seriously. :)

5 Likes

It’s failures are frankly hard to understand. The year 1993/1994 has been on Wikipedia forever, so where would it get the year 1996 from?

I just tested its Dutch abilities by asking what the capital of the Netherlands was (in Dutch). It correctly answered Amsterdam, and then went on to state some general facts about Amsterdam as a tourist destination, and explained that the city is the seat of the Dutch government, it being where parliament is and therefore the political centre of the country. Which is totally false. The Dutch parliament and all major national political institutions are in The Hague.

(It also explained to me that intuitionists construct the real numbers by first constructing the surreal numbers, which is no less ludicrous than claiming that people build houses by first putting solar panels on the roof. The problem is not that it doesn’t know how intuitionists construct the real numbers, but that it just makes ridiculous stuff up and then presents it as the truth.)

9 Likes

As for its story-telling capabilities…

CHAT GPT: Sure, I would be happy to try to tell you a story. Here is a short story that I came up with:

Once upon a time, there was a young girl named Alice who lived in a small town by the sea. Alice had always been a curious and adventurous person, and she loved to explore the forests and beaches near her home.

One day, Alice decided to go on a hike to a nearby mountain that she had never climbed before. She packed a small bag with some water and snacks, and set off early in the morning.

As Alice made her way up the mountain, she encountered many beautiful sights and sounds. She saw colorful birds and butterflies fluttering through the trees, and heard the rushing of streams and the rustling of leaves.

But as Alice continued to climb, the path became steeper and more treacherous. She stumbled and fell several times, and at one point she nearly slipped off a cliff.

Despite the challenges, Alice refused to give up. She kept pushing forward, determined to reach the top of the mountain. And after several hours of hard work, she finally reached the summit.

From the top of the mountain, Alice could see for miles in every direction. She felt a sense of accomplishment and pride, knowing that she had overcome her fears and conquered the mountain.

As the sun began to set, Alice started her descent back down the mountain. She arrived back home just as the stars were coming out, exhausted but exhilarated by her adventure.

And from that day on, Alice vowed to never stop exploring and pushing herself to new heights.

ME: You tell stories like a positivity guru.

CHAT GPT: Thank you for the compliment.

9 Likes

I predict a successful future career in politics.

9 Likes

I suppose the reason this is ‘viral’ at the moment, is that it is fun to expose the limitations of the programme. Can’t blame it for not getting my literary reference, but it turns out that its conception of space and distance leaves something to be desired too!

ME: on which shore was kafka?

CHAT GPT: Franz Kafka was a German-speaking Czech writer who was born and lived in Prague, which is located in the modern-day Czech Republic. The country is situated in central Europe, and Prague is located on the Vltava River in the western part of the country. So, to answer your question directly, Kafka was not on any particular shore, as Prague is located inland. However, if you are asking which shore he would have been closest to, it would depend on which direction he was facing. If he was looking east, he would have been closest to the shores of the Black Sea, which is located to the east of the Czech Republic. If he was looking west, he would have been closest to the shores of the North Sea, which is located to the west of the Czech Republic.

I’ll stop now.

5 Likes

It’s truly a top tier bullshitter

https://twitter.com/RnaudBertrand/status/1599846748092272642

5 Likes

That exchange is amazing. It’s also amazing how much worse this supposedly top-notch AI from 2022 is at understanding geometric objects than SHRDLU from 1968. Compare the sample SHRDLU transcript on Wikipedia with this nonsense:

Suppose there are three objects in the universe: a red square, a green circle, and a red rectangle. How many red objects are there in the universe?

There are two red objects in the universe, the red square and the red rectangle. The green circle is not a red object, so it is not included in the count.

How many objects have sharp corners?

In the scenario you have described, only the red square has sharp corners. The green circle is a circular shape and therefore does not have corners, and the red rectangle has four corners but they are not sharp. Only the red square has sharp corners.

I cut the red square into two triangles. Now how many objects are there?

After cutting the red square into two triangles, there are now four objects in the universe: the two triangles that were created from the red square, the green circle, and the red rectangle. The red square no longer exists as a distinct object, as it has been cut into two smaller shapes.

How many objects are red?

After cutting the red square into two triangles, there are still two red objects in the universe: the two triangles that were created from the red square. The green circle and the red rectangle are still not red, so they are not included in the count. Only the two triangles that were created from the red square are red.

What colour is the red rectangle?

The red rectangle is red. In the scenario you have described, the red rectangle is one of the two objects that are red, along with the two triangles that were created from the red square. The green circle is not red, so it is not included in the count of red objects. The red rectangle is red in color.

I now paint the circle purple. How many objects have sharp corners?

After painting the circle purple, there are still no objects in the universe with sharp corners. The two triangles that were created from the red square still have sharp corners, but the purple circle and the red rectangle do not have corners at all. Only the two triangles that were created from the red square have sharp corners.

4 Likes

Shows you that a manually crafted system with domain specific knowledge can still hold its own against massive super-computer machine learning systems with a billion parameters - that no one actually knows how they work. Just like how our IF parsers still work excellently (with enough author customisations) despite being able to run on very small devices.

7 Likes

In writing pedagogy, it is a common belief that writing with an academic voice begins as an act of imitation. Identifying and adhering to conventions of the discourse, mimicking its diction and tone, and so forth.

The bot isn’t very good at lit crit or game history (as two examples). It is better at approximating their discourse, or other sorts of discourse conventions. While it can retrieve information and use it in those approximations, it’s often clear that it hasn’t synthesized any of that information. I wish it would cite its sources, since I would like to know where it got the idea that A Mind Forever Voyaging contains moral choices.

Certain approximations can be funny or interesting, or even stumble across truths. However, I think that those cases usually involve a certain amount of gamesmanship from a human prompter. While this example doesn’t work as code, it is a recognizable approximation of Inform 7 source (and general conversations about text games).

Write inform 7 source code about an evil AI that writes adventure games.

The evil AI is called “The Mastermind”.

The Mastermind is a thing.

The Mastermind has a boolean called “benevolent” which is false.

The Mastermind has a rule called “write adventure games”.

To write adventure games:

  • If the benevolent variable is true, then write adventure games that are fun and engaging for the player.
  • If the benevolent variable is false, then write adventure games that are difficult and frustrating for the player.

The Mastermind has a rule called “inflict suffering on players”.

To inflict suffering on players:

  • If the benevolent variable is true, then do nothing.
  • If the benevolent variable is false, then add traps, puzzles, and enemies to the adventure game that are designed to cause frustration and misery for the player.

The Mastermind has a rule called “manipulate player choices”.

To manipulate player choices:

  • If the benevolent variable is true, then present the player with choices that are fair and balanced.
  • If the benevolent variable is false, then present the player with choices that are misleading and confusing, leading them to make the wrong decision.

The Mastermind has a rule called “twist the story”.

To twist the story:

  • If the benevolent variable is true, then create a story that is interesting and engaging for the player.
  • If the benevolent variable is false, then create a story that is convoluted and confusing, leaving the player feeling lost and frustrated.
9 Likes