Home
Creators
Search Recent Random
Posts
Search DMs Popular Hash Lookup Tags Random
Importer
Import FAQ
Register Favorites Keys Review DMs Logout
Community
Telegram ThePornDude
Home Creators Posts Import Register Favorites Logout

What we have in common with world-ending AI (Frank Lantz) (Patreon)

Published:
2025-06-13 19:39:14
Edited:
2025-10-10 05:44:34
Imported:
2026-06

Downloads

Content

This week on Post Games, “What we have in common with world-ending AI.”

Act 1: Why every AI superfan and skeptic should play Universal Paperclips. A review conversation with AJ Fillari, co-host of dotzip and producer of Into the Aether 

For Patreon subscribers, an intermission: 5 "clicker" games to play after Universal Paperclips

Act 2: An extended interview with Frank Lantz, the creator of Universal Paperclips about designing a game from the AI's POV

Act 3: The news of the week, including cool game art for a good cause

Also for Patreon subscribers: a recording of Frank Lantz’s famous speech about the trouble with modern video games

This episode is part one of a two-part series on what we can learn about AI and ourselves through video games. Next week, we'll continue with "How I AI-proofed my career after imagining the apocalypse."

(Image: Chris Plante via Team Asobi/SIE)

Thank you for subscribing to Post Games on Patreon! You keep the show alive, get tons of cool stuff like my video series Video Games Journalism 101, and can listen to new episodes early! 

Act 1: Universal Paperclips, a review

Universal Paperclips (Browser/iOS/Android)

Dotzip’s episode on Universal Paperclips

What future artificial intelligence will think of our puny human video games (Polygon)

The Unexpected Philosophical Depths of the Clicker Game Universal Paperclips (The New Yorker)

Our weird robot apocalypse: How paper clips could bring about the end of the world (Salon)

Nick Bostrom’s paperclips problem, explained (Wikipedia)

Act 2: Frank Lantz on designing the AI POV

Frank discusses his book, The Beauty of Games

“Most games are junk”

Advanced artificial intelligence will change everything, we are told. Do we get a say? (NYMag) 

Frank’s previous company, Area/Code, produced one of my favorite games: Drop7. Unfortunately, Zynga owns the game and has been a terrible steward of its iOS and Android updates.

Act 3: News of the Week

Pragmata is a game from a bygone era, and that era rules (Digital Trends)

Sonic Racing: CrossWorlds might be the most mechanically rich kart racer ever made (Eurogamer)

2025’s selected Game Maker’s Sketchbook art (iam8bit)

The Inescapable Intersection of SGF and ICE Protests (Pen to Pixels)

Just for Patreon: Five great clicker games to play after Universal Paperclips

The genre’s first breakaway hit: Cookie Clicker

A meta critique of clickers that is a clicker: Cow Clicker

What happens when you blend clickers with RPGs: Candy Box

Clicker game but make it trippy: The Gnorp Analogue

The modern evolution of the clicker formula: Cauldron

Free game of the week:

All Living Things demo: The Steam synopsis for All Living Things sounds promising enough: “Uncover the secrets of the philosopher’s stone by solving twelve intricately animated puzzles that reveal the mysterious hidden meanings within alchemical images.” But in the case of this claymation experiment, the visuals said everything I needed to hear.

The week in video game links

MindsEye From Ex-GTA Producer Is A Day-One Car Wreck (Kotaku)

'The game should speak for itself': MindsEye's publisher denies developer claims of a conspiratorial 'effort to trash' it (PC Gamer) 

“One of the most Sega games ever released” (Kimimi The Game-Eating She-Monster)

Striking SAG-AFTRA Voice Actors Reach Tentative Video Game Deal (Aftermath)

Baby Steps isn’t about humiliation, but the joy of disobedient flesh (RPS)

Drafting the Archaeological Mental Model of Blue Prince (Florence Smith Nicholls)

What else I'm watching

Love a short movie? Can’t get enough thrillers? Hate McCarthyism and believe that movies (and the artists who make them) have the power to inspire social change, however small it may be? Watch He Ran All the Way, a fantastic 77-minute, nail-biting drama that’s part of Criterion’s current series, Noir and the Black List.

Files

Comments

Groove Wizard

I'm still picking through the episode (end of act 2), writing this as I'm grinding for Trust in Universal Paperclips, but holy cow, a mountain of an episode is right. I love the WELL-EARNED confidence to grasp a topic such as *how capitalistic-driven generative AI intersects with art and the human condition* on the fifth episode of a video game podcast. Frank Lantz is a mad-lad, the kind of rare brain that you need to absolutely sprint to keep up with (which I thought you did a great job at!)

Alex Russell

Universal Paperclips is a fantastic and important game to play and that is 10,000% not what I expected the guy who made it to be like! I’m as deep an AI hater as it’s possible to be but I still enjoyed this episode. I’m impressed you were able to make something about AI that didn’t make me foam at the mouth. This show is great.

Daniel Gauthier

Such a great conversation with Frank. This show rocks

Directional Joy

Pragmata sounds a little Binary Domain to me, based off the thought that it looks, initially, average but may be looked at more fondly, in the future. I’ll keep my eye on it. 😃

Robby

Just listened, and I'm glad you touched on these topics! I teach on related material, and I thought maybe some context and further material might be of interest to other listeners. Bostrom gets a shoutout twice here, and he clearly is a huge influence on Frank, so I think it is worth mentioning that the story surrounding his work and his personal views is a bit problematic. He's really popular with tech bros, as they basically funded his Oxford center that he then ran into the ground, after he himself was called out for use of the N-word and saying white people were smarter than black people in a group email as a postgraduate. You might also argue that this sort of viewpoint is baked into the positions that came out of the Oxford center (longtermism and effective altruism) that many have suggested ought to be seen as a form of eugenics for the rich - consider that Musk was a benefactor of Bostrom. This Guardian piece is a nice discussion (https://www.theguardian.com/technology/2024/apr/28/nick-bostrom-controversial-future-of-humanity-institute-closure-longtermism-affective-altruism). Some other cool recent pieces that touch on points that came up - Hicks et al (https://link.springer.com/article/10.1007/s10676-024-09775-5) have a nice argument that we ought not to think of ChatGPT or other LLMs as aiming at truth at all, and so not as "hallucinating" or "making mistakes". Instead, since LLMs are just probabilistic next word predictors, we would do better to think of them as bullshitters, i.e., as similar to people who bullshit in the sense that they do not care about the truth at all. On the other hand, Buckner has an interesting response to similar criticisms that suggest we ought not compare LLMs to human cognition in that it reveals an anthropocentric bias (https://www.journals.uchicago.edu/doi/10.1086/714960).