After Things Podcast

Brian’s hodge-podge to-do system led to catastrophe. Smart devices and “accessing the home network.” An update on Andrew’s experience using Tonal, the home-gym installed on a wall. Send your project questions/ideas to [email protected], subject line “After Things.”

The episode opens with Andrew Mayne discussing OpenAI's newly released plugins and the broader idea of ChatGPT becoming more like an app platform. He describes practical examples such as Wolfram Alpha, Instacart, OpenTable, and browsing, then spends much of the conversation demonstrating Code Interpreter inside ChatGPT by generating mazes, GIFs, sound files, Game of Life music, QR codes, and even a simple chess engine. He frames these tools as making computation and creation more accessible to ordinary users. The discussion then shifts to everyday uses of ChatGPT and related AI tools: drafting letters, brainstorming, choosing meals, and helping with domestic decisions. Bryce introduces Rewind.ai, which records computer activity for later recall, leading to a privacy-and-utility discussion about AI-powered memory, search, and personal assistants. The episode also covers authenticity concerns around generated content and signatures, misinformation and viral falsehoods, the Hyperion tree, a disputed goblin shark photo, and a set of recommendations including Lucky Hank, Atlanta season three, Pinball: The Man Who Saved the Game, Screen Picks, and Riley Brown AI. Key topics ChatGPT plugins and app-like integrations: Andrew describes plugins as allowing ChatGPT to connect to tools like Wolfram Alpha, Instacart, OpenTable, and browsing so it can do more than plain text generation. Code Interpreter and computation inside ChatGPT: Andrew gives examples of running Python in ChatGPT to c

Bryce has an update on how his Blind Corners project went from 0 to 1 with a Substack. How do you break down barriers or build excitement? Send your project questions/ideas to [email protected], subject line “After Things.” Picks: Andrew: Arduboy Justin: ChatGPT-4 Brian: Ted Lasso season 3 Bryce: DALL-E

The episode is largely a discussion of ChatGPT/GPT-4 and how the hosts are using it in practice. Brian describes a hands-on experience role-playing as Blaine the Mono in GPT-4 and being impressed by how well it stayed in character, while Andrew explains the progression from GPT to GPT-2, GPT-3.5, ChatGPT, and GPT-4, emphasizing reinforcement learning with human feedback, instruction following, and stronger contextual understanding (L53-L61, L67-L69, L83-L85, L107-L118). The conversation also covers limitations and concerns: hallucinations and confident wrong answers, safety and guardrails, privacy/data-retention cautions, training data quality, and multimodal image features such as describing a refrigerator photo to generate recipes. The episode closes with picks for Welcome to the Monkey House, History of the World Part 2, Hello Tomorrow, Star Trek: Picard season 3, and a gel blaster/orbeez-style gadget discussed jokingly in the context of wildlife deterrence (L115-L118, L155-L177, L187-L209, L255-L269, L295-L297, L315-L317, L359-L365, L371-L405). Key topics ChatGPT roleplay and character emulation: Brian describes using GPT-4 to role-play as Blaine the Mono and says it answered in character and felt remarkably alive (L53-L61, L67-L69, L99-L101). OpenAI safety, guardrails, and misuse concerns: Andrew explains why the system has guardrails and safety policies, including avoiding harmful advice and misuse by bad actors or authoritarian governments (L83-L93). Instruction follow

Bryce has an idea, but not much more than that? What can a F1 pickem league turn into and how do you plus it? Plus, ChatGPT and a pitch for a presidential TV show. Send your project questions/ideas to [email protected], subject line “After Things.”

The episode opens with a long discussion of the Department of Energy report and other government assessments about COVID origins. The hosts say they do not know where COVID came from, but argue that lab-leak possibilities should not have been dismissed, and they criticize the shutdown of discussion on social media and within parts of the scientific community. The conversation widens into concerns about polarization, conflicts of interest, undisclosed funding, and the need for a serious investigation into how the origins question was handled. The show then moves through a set of lighter and more varied topics, including a Crew Dragon docking update, a conversation about SpaceX launch cadence, and a detailed discussion of Brightline rail fatalities and safety design in Florida. Later segments cover a story about video games for dogs, a creepy Reddit post about a house hidden inside an attic, and a Massachusetts school crawl space that may have concealed a crypto-mining setup. The episode ends with the usual picks, including recommendations for TV, a podcast, and a Jack London novel. Key topics Lab-leak theory versus zoonotic origin: The hosts contrast a lab-leak hypothesis with animal-origin explanations and repeatedly stress uncertainty. They argue the central issue is not certainty but whether the question was investigated seriously and fairly. Conflicts of interest and research transparency: Andrew raises concerns about undisclosed funding, research ties, and organizations s

Rovio announced a surprising change to one of their flagship games: they’re trying to remove it! Value propositions and how difficult it is to change directions once your project is out. Send your project questions/ideas to [email protected], subject line “After Things.” Picks: Justin: TV Funhouse – Disney Vault Brian: The Ur-Quan Masters Bryce: Schafpudel’s guide to Petz

The episode opens with a discussion of an Embraer patent application for an aircraft-seat system that scans passengers' faces, analyzes facial expressions, and may offer non-invasive transcranial stimulation. The hosts first frame it as a possible customer-satisfaction or unruly-passenger tool, then gradually settle on the idea that it could function as a passenger amenity for calming or helping people relax during flights, while also joking about the broader discomfort of air travel and the fantasy of skipping the flight entirely. A large middle section turns to AI-assisted communication. The hosts discuss Samsung's Bixby voice-cloning feature, then broaden into a near-future vision where calls, texts, voicemail, FaceTime, car play, and automated summaries all blur together. Later, they shift to media authenticity and trust, arguing that AI, image processing, and altered video will force society to rely on some kind of certification or verification system for human identity and evidence. Key topics AI-based passenger emotion tracking on airplanes: The Embraer patent includes a digital camera and facial-expression analysis to identify a passenger's emotion, with the hosts speculating about using it to measure customer satisfaction or detect when people are unhappy on a flight. Non-invasive transcranial stimulation as an in-flight amenity: The third part of the patent is described as offering distressed passengers non-invasive stimulation, including direct current, magnetic st

Adapting and adaptability. How do you balance quality and quantity while taking advantage of new spaces? Effort in a world where effort gets more and more efficient. Send your project questions/ideas to [email protected], subject line “After Things.” Picks: Andrew: Super Mario Odyssey Brian: Cunk on Earth Bryce: Poker Face

The episode opens with Andrew talking about how his work at OpenAI makes him more cautious about commenting on outside stories, then moves into a run of stories about elaborate pranks and public misdirection. Andrew recounts two balloon-based stunts from his past: putting a motorcycle inside a balloon for an A&E segment and, earlier, faking a UFO over a Florida school field with a lit balloon, which drew curious bystanders, radio calls, and a sheriff's helicopter before they fled and hid. The conversation then shifts to the long-running Hollandale, Florida outhouse prank, with the hosts reading a news account about an outhouse dropped at the post office every Halloween for years, sometimes in a custom White House shape. From there they discuss what makes a prank feel like performance art, compare old high-effort deception to modern prank videos, and talk about travel as an experience, including a British Rail tavern car and United's former men-only Chicago Executive flight. Later the episode covers a World War II bomb found in Yarmouth, England, then turns to the LAPD bomb squad incident in South Los Angeles where illegal fireworks were packed into a containment vessel despite warnings and the blast damaged vehicles, properties, and injured people. The final stretch becomes the picks segment, where the hosts discuss a book about flying cars and gyrocopters, Bryce recommends Midnight Suns, Brian recommends The Modern Rogue's thermite car episode, Andrew recommends The English,