The FAIK Files

Welcome back to The FAIK Files! In this week's episode: Google finally releases data on the power consumption and climate impact of its AI models A deep dive into 'Hierarchical Learning Models' (HRMs) and why they might be the next big thing in AI architecture How a smart home was reportedly hacked using a Google Gemini calendar summary A look at Perry's recent 'Deceptive Minds' episode, "The Long...

Welcome back to The FAIK Files! In this week's episode: OpenAI's GPT-5 launch that definitely didn't go according to plan, complete with backlash, safety concerns, and frantic updates Anthropic's fascinating new research on "persona vectors" - a breakthrough method for monitoring and controlling character traits in language models A listener tip leads us down a rabbit hole of ChatGPT conversations...

What does it mean to think like a hacker? In this episode, we're joined by Ted Harrington — ethical hacker, speaker, and executive partner at Independent Security Evaluators, where he leads a team that breaks things on purpose to make them stronger. We talk about the hacker mindset: how curiosity, exploration, and questioning assumptions can lead to stronger systems — and more creative thinking. T...

Welcome back to The FAIK Files! In this week's episode: NVIDIA’s Diffusion Renderer: We explore NVIDIA's new research into real-time diffusion rendering, a potential game-changer in photorealistic generation Robotics Powered by Video AI: Luma and Runway are targeting robotics companies with video synthesis tools—because training a robot brain takes a lot of fake video AI-Generated Fashion Models:...

Welcome back to The FAIK Files! In this week's episode: Subliminal Learning: Listener "Max" brings us a fascinating (and concerning) story about AI model distillation and how "dark knowledge" can be passed along between AI systems AI Filmmaking vs. Fraud Crisis: We explore Dave Clark's innovative AI agent methodology for filmmaking, then pivot to Sam Altman's growing concerns about AI's potential...

Welcome back to The FAIK Files! In this week's episode: Grok's triple threat of chaos - Elon Musk's AI chatbot continues identifying as Hitler, starts generating anime waifus, and somehow lands a $200 million Department of Defense contract ChatGPT's accidental generosity - A seemingly innocent guessing game vulnerability leads users to extract free Windows OS keys and other unintended system acces...

Welcome back to The FAIK Files! In this week's episode: Meet Claude, Small Business Owner Tricking AI – two different ways Shifting Policies Around AI Content Dumpster Fire of the week: Grok goes full Nazi Check out Perry's new website, ⁠⁠⁠⁠⁠⁠⁠The Deception Project⁠⁠⁠⁠⁠⁠⁠ and newsletter, Deceptive Minds: a newsletter about how we are fooled, how we fool ourselves, and what we can do about it. Subs...

Welcome back to The FAIK Files! This week, we're sharing our interview with Winn Schwartau! Winn Schwartau is a respected authority in information security and one of the earliest voices to warn about the risks of cyberwarfare. With a career spanning decades, he’s written extensively on topics like critical infrastructure vulnerabilities, time-based security, and the human factors behind digital t...

Welcome back to The FAIK Files! In this week's episode: Does ChatGPT make us dumber? AI chatbots are amplifying disinformation Anthropic's recent "fair use" court victory and what it actually means Dumpster fire of the week: ChatGPT babysitters and therapy chatbots Check out Perry's new website, ⁠⁠⁠⁠⁠⁠⁠The Deception Project⁠⁠⁠⁠⁠⁠⁠ and newsletter, Deceptive Minds: a newsletter about how we are fool...

This week, there' s no real episode! Instead, please enjoy this song-summary of what we had planned to cover today. If you want to dig in on your own, here's the sources we were planning to talk about: SEGMENT 1: YOUR BRAIN ON CHAT GPT Does LLM use build up "cognitive debt" and make you dumber? (Probably not TBH) Media page from the authors of the study: https://www.brainonllm.com The study itself...

Welcome back to The FAIK Files! In this week's episode: OpenAI's models are reportedly resisting human commands to shut down A controversial new research paper questions whether AI systems actually "think" or just create an illusion of reasoning. How generative AI exploits the same psychological triggers as gambling—"just one more prompt, bro!" Google's VEO-3 and ElevenLabs 3 are creating deepfake...

Welcome back to The FAIK Files! In this week's episode: Perry is on some kind of secret government mission??? Mason showcases a cool AI art project, and some behind-the-scenes processes of integrating AI elements into art. We explore some more eye-rolling and stupid uses of AI art, including in scientific publishing and the music industry And we interview Chris Machowski, artist and overall creati...

Welcome back to The FAIK Files! In this week's episode: Google wants to help people speak dolphin... and they have a new language model to prove it! Perry brings a couple fun headlines... And we interview Paul Vann and Justin Marciano of Validia.ai Check out Perry's new website,  ⁠⁠⁠⁠The Deception Project⁠⁠⁠⁠ and newsletter, Deceptive Minds: a newsletter about how we are fooled, how we fool oursel...

Welcome back to The FAIK Files! In this week's episode: Quantum computing is accelerating fast Professors staffed a fake company entirely with AI agents (spoiler: chaos) Microsoft is adding native AI agent support to Windows - what could go wrong? The Chicago Sun-Times accidentally printed an AI-generated summer reading list with books that don't exist Check out ⁠⁠⁠The Deception Project⁠⁠⁠ to lear...

Welcome back to The FAIK Files! In this week's episode: An Australian radio station created a fake Asian female host using AI The BBC resurrects Agatha Christie while a family member brings a murder victim to court We break down "Strategic Text Strings" - sequences of gibberish that can jailbreak AI systems AI recruitment tools might be making hiring worse, not better Check out ⁠⁠The Deception Pro...

Welcome back to The FAIK Files! In this week's episode: That AI bot may be an undercover AI cop Playing with Suno's "Cover" Feature Bad actors using Claude to orchestrate disinfo campaigns Does being honest about your AI use make people trust you less? AI companies are looking to cash-in on human loneliness Check out ⁠The Deception Project⁠ to learn about our upcoming ⁠Offensive Cyber Deception Ma...

Welcome back to The FAIK Files! In this week's episode: We explore the One Million Chessboards project, a massive collaborative web experiment where users can move pieces across a million shared chessboards in real-time Anthropic's model welfare research program, AI ethics, and the need for interpretability OpenAI's recent struggle with ChatGPT's personality crisis as they roll back an update that...

This is the 2nd of two sample episodes/issues of Perry's audio newsletter. If you enjoy these and want to keep listening, go ahead and subscribe here. ----- Hey friends! Here's issue 2 of Deceptive Minds: a newsletter about how we are fooled, how we fool ourselves, and what we can do about it. 🔥 Since starting this newsletter last week, we're already closing in on 3,000 subscribers to the ⁠print...

 Hey listeners, Perry Carpenter here. I've got something exciting that I think you'll be interested in. It's the audio version of a newsletter that I recently started. The newsletter is called Deceptive Minds, and it's all about how we are fooled, how we fool ourselves, and what we can do about it. I'm dropping the first two episodes/issues in this feed so you can decide if you like it and might b...

Welcome back to The FAIK Files! In this week's episode: Why being polite to ChatGPT might be wasting Sam Altman's money Alarming advancements in deepfake technology and why even "super recognizers" can't keep up. We explain what a "vibe coder" is and why you probably don't want to become one. An Italian newspaper's claim to have published the world's first AI-generated edition. Check out The Decep...

Welcome back to The FAIK Files! In this week's episode: Mason talks about AI powered spreadsheets! ... it's cooler than it sounds Perry & Mason cover a few fascinating AI headlines that caught our attention this week Mason takes us on a journey into musician Benn Jordan's innovative "Poisonify" project that disrupts AI training on music And our AI Dumpster Fire of the Week: AI Chatbots that will c...

Subscribe to Me, Myself, and AI: https://sloanreview.mit.edu/audio-series/me-myself-and-ai/ Today we're sharing with you an episode of another show we like called Me, Myself, and AI. It's a podcast produced by MIT Sloan Management Review and Boston Consulting Group, where hosts Sam and Shervin interview various folks from all sorts of industries and walks of life about the constantly evolving land...

Welcome back to the FAIK Files! In this episode: Mason fills in for Perry as best he can, with a mix of topics that all fall under the broad umbrella of "functionally useless, but cool". Well... except the last segment; that one is inarguably kind of un-cool. We play a little bit of a 100% AI-generated rendition of Quake 2, and discuss how it works under the hood - as well as the general reaction...

Welcome back to the FAIK Files! In this week's episode: We explore "Prompt Wars," an asynchronous multiplayer LLM competition where players craft prompts to battle against others in a fascinating game of AI manipulation We dive into Anthropic's latest research on AI cognition and circuit tracing - essentially digital neuroscience for large language models Mason shares his AI BodgeLab experiment cr...

Welcome back to the FAIK Files! In this episode: Perry cracks open a chaotic grab bag: A couple updates from Perry and the National Cybersecurity Alliance. Plus two new reality-bending image models that raise the bar again — this time with people that really look like people and text that actually looks like text! We sit down with former FBI special agent, Cameron Malin.— a deception expert and be...