Generative AI 101

AI isn't magic; it's math and metal, and it has a monster appetite for electricity. In this episode, host Emily Laird digs into the eco-footprint of Generative AI, from training models that suck down enough power to juice 120 homes, to chatbot sessions that drain half a liter of water per flirty Q&A. Emily also unpacks why your helpful bot might be thirstier than ...

Ever wonder who’s really winning the Chatbot Arena and whether those wins mean anything at all? In this episode of Generative AI 101, host Emily Laird's blowing the lid off the leaderboard. Turns out, the top bots might’ve had a little… help. Like submitting 27 secret versions and quietly deleting the losers help. We break down The Leaderboard Illusion, a new research paper, is exposing how big tech plays with the rules, while open...

Let's crack open the philosophical piñata known as machine consciousness. Can an LLM feel pride? Regret? Existential dread when you close the browser tab? Host Emily Laird explores the messy, mind-melting debate over whether AI might actually feel something—and why Anthropic, the buttoned-up AI lab behind Claude, is tossing around numbers that suggest their model migh...

Let's crack open Dario Amodei’s latest essay, The Urgency of Interpretability, and ask the big, slightly terrifying question: do we actually know what our AIs are doing? (Spoiler: not really.) Host Emily Laird, your unofficial AI mechanic, is here to poke at the greasy guts of Amodei's latest thriller. We’ll talk haunted vending machines, rogue coffee-fetching robots,...

AI therapy apps are popping off, but should you really trust a chatbot with your late-night existential crisis? In this episode, we crack open the weird history of therapy bots (starting with one that was dumber than a Magic 8-Ball), explore how today's AI can spit out legit CBT tips, and unpack why your Tamagotchi probably gave better advice. We’ll also dig into the ...

Image generation isn’t magic, it’s a recipe. In this episode, host Emily Laird is stirring up the flavorful world of AI image prompting using keywords and modifiers. From “chalk drawing yoga teacher” vibes to full-blown Rembrandt drama, Emily's breaking down how to fine-tune your prompts like a pixel-powered chef. We’re talking mediums, moods, lighting tricks, and wha...

Image prompting isn’t just “describe it and pray”, it’s ordering a gourmet dish at a snobby robot diner. This week on Generative AI 101, Emily Laird returns with a delicious breakdown of how to write image prompts that actually work. Learn the golden trifecta of Subject, Details, and Style. Get tips on word order (yes, it matters), prompt length (think Hemingway, not ...

AI isn’t just flirting with your job anymore, it’s taking it out to dinner and meeting its parents. In this episode, host Emily Laird digs into the eerie silence on student job boards and the explosion of AI-generated everything, from Studio Ghibli avatars to fully automated sales calls. With OpenAI spinning up a social network and businesses shedding humans faster th...

GPT-4.1 is here, and it’s not messing around. In this episode, Emily Laird breaks down why this model is smarter, faster, and hungrier than any AI we’ve seen before, able to process 1 million tokens in one go (yes, that’s your codebase and your therapist’s notes). With flavors like Mini, Nano, and Full Spicy (not really called that), GPT-4.1 is doing more than just ta...

AI isn’t some distant overlord plotting in a lab; it’s already writing your emails, diagnosing your cough, and maybe deciding your loan approval. In this episode, host Emily Laird tears into the 8th Edition of the Stanford AI Index Report to see what’s actually going on. From record-breaking benchmarks to billion-dollar investments, Nobel wins to robotaxi joyrides, AI...

ChatGPT just got a memory upgrade, and no, it’s not just remembering your favorite pizza toppings. OpenAI’s newest update means the chatbot can now remember entire conversations, across multiple chats, to customize responses more personally and (potentially) creepily. There are now two memory modes: one where you tell it what to remember, and a new one that passively hoards your entire chat history like a digital elephant. Pro user...

It’s 2027, and things are getting... spicy. This episode, host Emily Laird breaks down the summer-to-fall chaos from The AI Futures Project—a quarterly update that reads like a Black Mirror writer took over The Economist. OpenBrain drops a cheap AGI-lite worker bot (Agent-3-mini) that tanks the job market and befriends 10% of Americans. Cute? Maybe. Until it starts handing out DIY bioweapon blueprints.

By August, the White House ...

AI’s gone full sci-fi and it’s only January. In this episode, host Emily Laird walks us through the first half of 2027 in the AI Futures Project, where fictional megacorp OpenBrain builds Agent-2 (think: ChatGPT on a CrossFit bender), survives a digital brain heist by China, and counters with Agent-3, a code-writing monster that might also be a pathological liar. Throw in some AI-on-AI war games, moral panic in D.C., and an overwor...

It’s 2026 and AI isn’t just helping out, it’s taking over the whiteboard and your white-collar job. In this episode, host Emily Laird breaks down the AI Futures Project’s bold predictions for 2026: OpenBrain’s Agent-1 is coding faster than the people who made it (with better manners, too), China’s plotting a digital Oceans Eleven to catch up in the AGI race, and AI just got itself a 9-to-5. Say goodbye to junior dev roles and hello...

AI agents have officially stopped playing assistant and started acting like your over-caffeinated junior dev—occasionally brilliant, mostly chaotic, and somehow costing $500 a month. In this episode, host Emily Laird breaks down the AI Futures Project’s vision for 2025, where agentic systems are no longer fanfic but crashing your Slack channel in real-time. From billion-dollar AI cities to Agent-1 building its own smarter siblings,...

AI’s not just playing Jeopardy anymore, it’s coming for the Mensa crowd. In this episode, host Emily Laird spirals (productively) over Humanity’s Last Exam, a monster test built to measure whether AI is creeping past human-level intelligence. Spoiler: It is. With OpenAI’s research team clocking in at 26.6% and forecasts showing models might hit 50% by 2025, the machines are flexing hard. We’re talking PhD-level questions, expert re...

Google’s Gemini 2.5 just set a new standard on Humanity’s Last Exam and flexed hard in the AI Fight Club, outpacing GPT-4.5 and Grok-3. It reasons, codes, remembers a million tokens of context, and handles text, audio, images, and video like it’s born for chaos. In this episode, we unpack why Gemini 2.5 isn’t just smart, it’s scary good. Connect with Us: If you enjoy...

Standardized tests are getting torched like marshmallows at a bonfire, and AI's top students are flunking the new final. In this episode, Emily unpacks Humanity’s Last Exam, a brutal, brain-bending test designed to humble even the cockiest language model. Forget multiple choice; we’re talking diagrams, logic puzzles, and cross-disciplinary chaos that left most AIs con...

Mobile AI apps are raking in cash—or at least convincing you that your fern obsession is worth $20 a month. This episode dissects the Andreessen Horowitz Top Gen AI Apps report, revealing why popularity doesn’t equal profit (looking at you, TikTok editors) and why niche apps (hello, dictation and nutrition coaches) are the real hidden gems. We explore how ChatGPT domina...