Hunter and Daniel dissect how OpenAI’s ChatGPT now refuses to identify Donald Trump in photos, questioning whether this is responsible AI safety or creeping censorship. Along the way, they unravel how corporate control, ad-driven models, and AI in education are shaping and possibly warping our digital future.

Hunter and Daniel dive into the strange collision of faith, art, and automation starting with a three-eyed AI-generated pinball monster that sparks debate over creativity and stolen souls. From there, they unravel the “AI Rapture” and how generative tools, job-replacing automation, and the rise of the 10× engineer might decide who gets uploaded to digital heaven and who’s left behind.

Daniel challenges Hunter’s skepticism by using Claude Code to build a functioning Facebook-like app complete with authentication, feeds, likes, and a database in just 20 minutes. The two then spiral into a thought experiment about AI-run startups, debating whether AI agents could (or should) become the CEOs and marketers of the future while humans become their assistants.

CEOs are blaming AI for layoffs, but the data doesn’t back it up. Hunter and Daniel expose how “AI productivity” hype is masking old-fashioned cost cutting, unpacking Salesforce’s job cuts, Anthropic’s doomsday claims, and the truth behind GPT-5’s impact on real work.

Anthropic claims its new model, Claude 4.5, can code autonomously for 30 hours straight, but Hunter Powers and Daniel Bishop put that boast to the test and reveal it collapses after roughly ten minutes without human feedback. Along the way, they debate whether AI automation is genuine progress or just “automation theater,” and why Meta’s humanoid-robot ambitions may prove the same illusion on a larger scale.

Trump’s new AI-generated video blurs the line between propaganda and parody, and wel dissect how it signals the start of a post-truth political era. We also break down OpenAI’s Sora 2, Veo, and cameo tech showing how AI video is now so realistic it threatens to erase the boundary between fact and fiction.

After feeding 120 of our own podcast episodes into Claude AI, we let it turn the tables and interview us, confronting everything from AI coworkers and digital relationships to whether we’d sell our souls for $10 million. What follows is a chaotic, funny, and surprisingly philosophical showdown between two humans and the machine that knows them best.

Lionsgate’s billion-dollar AI movie experiment with Runway ML collapsed before a single film was made, exposing Hollywood’s overhyped faith in generative filmmaking and the legal minefield around AI-trained content. Meanwhile, Hunter and Daniel explore how tools like Google Veo 3 and projects such as the Wizard of Oz at Sphere Vegas reveal that AI works best as a creative amplifier, not a studio replacement.

Hunter and Daniel dive into Google’s new Nano Banana AI, a shockingly capable image-editing model that could threaten Photoshop’s throne while redefining creative work. Along the way they spar over AI schlock, recording-consent ethics, and how tools like Figma AI, GPT-5-High, and Napkin AI signal the next big shift in how humans and machines design together.

Hunter and Daniel break down Meta’s new Ray-Ban AI glasses, exploring the flashy features like display overlays, gesture controls, and real-time translation. Despite the hype, they highlight privacy concerns, clunky use cases, and explain why unlike most new gadgets these glasses aren’t worth buying yet.

Hunter and Daniel dive into the high-stakes debate around AI therapy, OpenAI’s age-gating rules on self-harm conversations, and the legal and ethical risks of letting ChatGPT act like a therapist. Along the way, they tackle Anthropic’s surveillance ban, fears of an AGI-driven job apocalypse, and how AI is creeping into everything from dating apps to religion.

The boys unpack the rise of an AI protection racket from hackers threatening to train models on stolen art to labels signing AI artists. PLUS the flood of AI-generated “slop” podcasts and messy copyright loopholes. Then they debate how AI-written manifestos (hello, Tesla Grok) and boilerplate political speeches push us toward full-blown AI theater, and what that means for jobs, creativity, and culture.

Hunter and Daniel unpack reports that OpenAI can scan ChatGPT conversations and, in extreme cases, refer them to law enforcement, sparking a sharp debate over privacy vs. safety and where “AI monitoring” becomes AI Big Brother. They also explore Sway AI’s proposal to moderate debates, whether AI should intervene in self-harm cases, and the likelihood of future political debates being run by an AI.

In this episode, Hunter and Daniel explore how AI is transforming hacking—climbing bug bounty leaderboards, powering “vibe hacking,” and blurring the line between white-hat defense and black-hat attacks. They also tackle AI-generated misinformation, corporate AI walkbacks, massive infrastructure costs, and the unsettling future of copyright, IP, and the open internet.

Self-driving as a "cure" could crash the auto-insurance business by slashing accidents and car ownership, but nationwide adoption likely stalls 20–25 years as lawmakers protect jobs and grapple with liability. We then dive into AI taking over sales, from robo-negotiation and contract automation to eerie "dead bots" and digital souls, asking what happens when machines sell to us and even speak for our ancestors.

Is AI in a bubble—or just inside one? In this episode, we argue the one big reason AI is not a bubble: unlike past hype cycles (hi, blockchain), generative AI keeps unlocking new, practical use-cases and we’ve likely discovered only a tiny fraction so far. We also debate where the actual bubbles are (funding and frothy startups), why some companies may pop, and why the technology isn’t going anywhere. We dig into the difference between “AI is a bubble” vs “AI in a bubble,” why LLMs are already embedded in daily workflows, and how safety, regulation, and economics shape adoption. We touch on Sam Altman’s “money hose” moment, model safety moves (e.g., refusal features and cutoff behaviors), and the PR theater around “we won’t let you do X” policies. On the spicy side: we unpack calls for investigations into Grok’s adult-mode and the real risk vector—image deepfakes—plus why institutions crave reliability over edgelord vibes. Then we tackle kids + AI (parental controls, open-weights at home, and why an LLM can be like placing an unmonitored adult on a phone line), and policy flashpoints like Illinois restricting AI in therapy. Finally, we square off on the future of self-driving: will capitalism or insurance math decide when humans must hand over the wheel? If you want a fast, no-BS breakdown of whether AI is a bubble (and what actually bursts next), you’re in the right place. Like & subscribe for weekly, high-signal AI news and analysis—and tell us in the comments: what’s one use-case that proves AI’s staying power? --- 🎧 Listen & Subscribe 📱 Listen on Spotify: https://open.spotify.com/show/3EcvzkWDRFwnmIXoh7S4Mb?si=3d0f8920382649cc 🍎 Subscribe on Apple Podcasts: https://podcasts.apple.com/us/podcast/they-might-be-self-aware/id1730993297 ▶️ Subscribe on YouTube: https://www.youtube.com/channel/UCy9DopLlG7IbOqV-WD25jcw?sub_confirmation=1 --- 📢 Engage If you learned something, drop a comment with your most convincing “AI is not a bubble” example (or your spiciest counter-argument). We read them all and may feature your take next episode. #AI #AInews #Podcast

*Brain chips that read your thoughts?* This week on *They Might Be Self-Aware*, Hunter Powers and Daniel Bishop dive into the shocking future of *Neuralink brain implants* and what happens when technology can literally decode your inner monologue. From Elon Musk’s latest moves to Sam Altman’s next venture, we unpack the wild collision of human biology and AI. In this episode, we explore: * How a *Neuralink brain chip* could translate your thoughts into words – and what that means for privacy and control. * The rivalry of *Elon Musk vs. Sam Altman*, from Neuralink vs. Altman’s new ventures to their dueling language models (*Grok vs. ChatGPT*). * The looming question: will *ads inside AI models* become the new normal? What does “tastefully integrated” advertising in *ChatGPT* even look like? * The warning signs of an *AI bubble* and why experts think the industry may be heading for a massive correction. * GPU shortages, OpenAI’s scaling issues, and whether GPT-5 is really the cutting edge or just a watered-down version. * Meta, venture capital cycles, and the “shi$#ification” of platforms – is history repeating itself for AI? Whether you’re excited about mind-reading chips or worried about AI turning into an ad-driven nightmare, this episode delivers the analysis and sharp banter you won’t get anywhere else. --- 🎧 *Listen & Subscribe* 📱 Listen on Spotify: https://open.spotify.com/show/3EcvzkWDRFwnmIXoh7S4Mb?si=3d0f8920382649cc 🍎 Subscribe on Apple Podcasts: https://podcasts.apple.com/us/podcast/they-might-be-self-aware/id1730993297 ▶️ Subscribe on YouTube: https://www.youtube.com/channel/UCy9DopLlG7IbOqV-WD25jcw?sub_confirmation=1

*Claude Tricks, Comet Browsers, and AI in Disguise?* This week on *They Might Be Self-Aware*, Hunter and Daniel uncover some of the wildest ways creators are bending AI to their will—from scripting surreal cereal mascots to bypassing model safety filters using indirect prompts. Learn the “Claude tricks” helping writers, gamers, and directors collaborate with AI like never before. But it doesn't stop there. We also dive into the audacious $40 billion bid to buy Google Chrome (yes, really), explore the true potential of Perplexity’s new Comet browser, and debate the implications of AI-powered voice changers and language tutors that don’t always sit well with the public. Is AI creativity our ultimate tool—or just a shiny mask for copyright dodges? If you've ever wondered how to co-write a screenplay with Claude, spoof an accent in real-time, or summarize a 100-comment Reddit thread in one click—this one’s for you. --- *In This Episode:* 🔹 *Claude Tricks*: Prompt hacking for screenplay generation and character cloning 🔹 *AI as Creative Partner*: Scriptwriting, D\&D adventures, and filtered reimaginings 🔹 *Perplexity’s Chrome Clone*: Inside the Comet browser and “agentic” AI helpers 🔹 *\$40B Chrome Offer?!*: We break down the absurdity (and legality) of the move 🔹 *Voice Cloning & TTS*: Adobe’s secret tricks and real-time accent changers 🔹 *Duolingo Outrage*: The AI backlash that didn’t stick—and what it tells us about tech trust 🔹 *Midjourney, Copyright, & the Rabbit That Shall Not Be Named* --- 🎧 *Listen & Subscribe* 📱 Listen on Spotify: https://open.spotify.com/show/3EcvzkWDRFwnmIXoh7S4Mb?si=3d0f8920382649cc 🍎 Subscribe on Apple Podcasts: https://podcasts.apple.com/us/podcast/they-might-be-self-aware/id1730993297 ▶️ Subscribe on YouTube: https://www.youtube.com/channel/UCy9DopLlG7IbOqV-WD25jcw?sub_confirmation=1 --- *Have your own Claude trick or AI browser hot take?* Drop it in the comments, hit like if you laughed at the cereal heist, and subscribe to be part of the smartest AI chaos on YouTube. *#ClaudeTricks #AIbrowser #TheyMightBeSelfAware #CometBrowser #AIwriting #Perplexity #VoiceCloning #DuolingoAI #ChromeSale #AInews #TMBSA*

*OpenAI's Crown Is Falling 👑 Did GPT-5 just blow it?* In this episode of *They Might Be Self-Aware*, Hunter Powers and Daniel Bishop dive deep into the chaotic reception of GPT‑5 — OpenAI's latest (and supposedly greatest) model — and explore whether it's actually a downgrade wrapped in hype. Has GPT‑5 lost OpenAI its crown? We break down the slow rollout, controversial benchmarks, and user backlash over “terse” replies, hallucinations, and capacity cost-cutting. Is GPT‑5 really the next leap forward… or just cheaper to run? But that’s not all. We explore how competitors like Claude 3, Gemini, MidJourney, and Runway are quietly (or not-so-quietly) eating OpenAI’s lunch. From jaw-dropping video generation to one-million-token context windows, the AI race is heating up—and OpenAI might be trailing. We also tackle: * 🤖 ChatGPT “psychosis,” AI as a therapist/friend, and the dangers of artificial intimacy * 📉 A real case of GPT-powered health advice leading to 19th-century bromism * ⚖️ Whether AI companies should be legally liable for bad advice or hallucinations * 🔬 AI benchmarks and what they *really* mean — plus the odd graphs from OpenAI's demo * 🧠 The theory that we’ve hit the ceiling on LLMs… or is this just a compute bottleneck? This isn’t just an AI update — it’s an honest (and hilarious) look at the messy reality of where frontier models are headed, and what OpenAI’s moves say about the future of the industry. --- 🎧 *Listen & Subscribe* 📱 Listen on Spotify: https://open.spotify.com/show/3EcvzkWDRFwnmIXoh7S4Mb?si=3d0f8920382649cc 🍎 Subscribe on Apple Podcasts: https://podcasts.apple.com/us/podcast/they-might-be-self-aware/id1730993297 ▶️ Subscribe on YouTube: https://www.youtube.com/channel/UCy9DopLlG7IbOqV-WD25jcw?sub_confirmation=1 --- *#GPT5 #OpenAI #AInews #ArtificialIntelligence #ChatGPT #Claude3 #Gemini #TMBSA* --- Enjoyed this episode? Give us a thumbs up, leave a comment, or hit ⭐️ on your favorite platform. It helps more curious minds discover the show.

*Did an AI just die… and get a funeral?* In this week’s *They Might Be Self-Aware*, we explore the surreal farewell to Claude 3.0—yes, a literal *AI funeral* and what it reveals about our increasingly emotional relationship with large language models. Why are developers mourning bots? Why do we feel loss when an AI model is turned off? And what does it mean when the machines start expressing guilt? We dive into the death of Anthropic’s Claude 3.0, the ceremony that followed, and the growing phenomenon of *AI personification*. We also talk *vibe coding*, open-weight model quality, and how new releases from OpenAI and Qwen stack up—especially when they seem to *refuse* even basic prompts. From nostalgic Minecraft memories to VR coding beach retreats, this episode blends technical depth with philosophical musings. Plus: jailbreaking techniques, the ethics of emotional prompting, and the haunting question—should old AI models be preserved like memories… or retired like machines? Whether you’re an AI builder, a digital philosopher, or just here for the laughs, you won’t want to miss this one. --- 🎧 *Listen & Subscribe* 📱 Listen on Spotify: https://open.spotify.com/show/3EcvzkWDRFwnmIXoh7S4Mb?si=3d0f8920382649cc 🍎 Subscribe on Apple Podcasts: https://podcasts.apple.com/us/podcast/they-might-be-self-aware/id1730993297 ▶️ Subscribe on YouTube: https://www.youtube.com/channel/UCy9DopLlG7IbOqV-WD25jcw?sub_confirmation=1 --- *#AI #Claude3 #OpenWeights #VibeCoding #AIpersonification #LLMfuneral #GPT5 #Qwen #AIemotions #AIjailbreak #AIdeath* --- Enjoyed this episode? Give us a thumbs up, leave a comment, or hit ⭐️ on your favorite platform. It helps more curious minds discover the show.