Topic section
Models and labs
4 topics and 32 episodes whose main topic belongs to this section, by topic, newest first. The companies come from the same episodes.
New models and comparisons · 12 episodes
GPT-6 Astra — A New Leap for AI? Where It Has Become 2–3 Times Stronger67% of the episode · Astra · OpenAI · FableWatch →AI Is Changing the Internet: What Happens to Browsers, Google, and Apps?48% of the episode · Claude · Google · Kimi K3Watch →ChatGPT, Claude, Gemini, and Grok Do More Than Answer Differently—Their Companies Give Them Different Political Characters92% of the episode · Gemini · ChatGPT · OpenAIWatch →ChatGPT-5.5 Reads Prompts and Gets More Expensive: Personalization Is Growing Faster Than Trust31% of the episode · ChatGPT · Grok · GPT-5.5Watch →GPT-5.4 Is Stronger, but Google Wins Where the Model Already Lives Inside the Documents25% of the episode · OpenAI · Google · AnthropicWatch →OpenAI and Anthropic Are Beginning a War of Business Models, Not Models: Advertising vs. Expensive Work27% of the episode · OpenAI · Anthropic · ChatGPTWatch →Gemini 3 Won a Noisy Week, but the Real Race Is for an Agent That Can Work, Not Take an Exam36% of the episode · Gemini · Grok · Elon MuskWatch →GPT-5 Became Easier to Use, but the Main Progress Is Not One Benchmark—It Is How the Model Handles a Task52% of the episode · OpenAI · GPT-5 · ChatGPTWatch →Grok 4 Is Impressive in Speed and Ambition, but a Work Tool Is Not Defined by a Benchmark50% of the episode · Grok · OpenAI · Grok 4Watch →o3 Unified OpenAI’s Tools but Did Not Solve the Main Problem—Confident Hallucinations47% of the episode · OpenAI o3 · OpenAI · ChinaWatch →Meta and Microsoft Are Dividing User Habits, Not Models30% of the episode · Microsoft · OpenAI · Deep ResearchWatch →Grok 3 Did Not End the ChatGPT Era—It Showed That the Race Now Runs Every Day25% of the episode · OpenAI · Elon Musk · GrokWatch →
The AI lab race · 6 episodes
AI Has Started Building Itself: Development Speeds Up, and the Main Human Skill Shifts to Choosing the Direction32% of the episode · Anthropic · OpenAI · Artificial intelligenceWatch →“Code Red” at OpenAI: Google Applies Product Pressure While Social Programs Buy Time and Influence25% of the episode · OpenAI · Google · NVIDIAWatch →OpenAI Is Building a Labor Market, DeepSeek Is Preparing a New Model, and Compute Scarcity Is Becoming the Main Employer28% of the episode · OpenAI · DeepSeek · Artificial intelligenceWatch →Waymo Already Looks Like Transportation, Tesla Like a Bet on Scale, and Meta Is Buying People at Company-Level Prices26% of the episode · Tesla · OpenAI · MetaWatch →OpenAI Is Not Finished, but One Strong ChatGPT Is No Longer Enough45% of the episode · OpenAI · ChatGPT · Artificial intelligenceWatch →ChatGPT Is Already Useful in Everyday Life, but the Industry Still Cannot Turn AI Into a Proper Product24% of the episode · OpenAI · Tesla · ChatGPTWatch →
How models work · 13 episodes
Has the Singularity Already Begun? Who Will Control Superintelligence?80% of the episode · Sam Altman · Technological singularity · OpenAIWatch →AI Is Accelerating Its Own Development: AGI Will Come Not From One Breakthrough, but From a Loop in Which Each Model Helps Build the Next61% of the episode · Artificial general intelligence · Anthropic · Dario AmodeiWatch →Google Is Pulling Ahead Because Personal AI Sees an Entire Digital Life, Not One Chat32% of the episode · Google · Artificial intelligence · Artificial general intelligenceWatch →Your Data Is Already Working for AI: A Wristband, TikTok, Meta, and Deepfakes Are Merging Into One Personal Profile24% of the episode · TikTok · ChatGPT · MetaWatch →There Is No Single Strongest Model: ChatGPT Has to Be Chosen Again for Every Task37% of the episode · OpenAI o3 · Google · OpenAIWatch →Word2Vec Does Not Read a Sentence—It Learns to Guess a Word From Its Neighbors100% of the episode · Word2VecWatch →How AI Turns Words Into Coordinates and Begins to See Relationships Between Meanings100% of the episode · Word2VecWatch →A “Think” Button Does Not Make a Model Conscious—It Shows the Cost of a Long Task32% of the episode · OpenAI · Deep Research · ClaudeWatch →After DeepSeek, the Market Started Asking Not Where a Model Comes From, but Where Its Output Can Be Trusted38% of the episode · DeepSeek · OpenAI · Deep ResearchWatch →The Data Is Running Out, and AI Has to Learn From Itself47% of the episode · Artificial intelligence · OpenAI · GoogleWatch →AGI Is Promised for Tomorrow, but No One Has Explained What Is Supposed to Arrive37% of the episode · Elon Musk · Artificial general intelligence · Sam AltmanWatch →OpenAI Taught a Model to Think Longer—but Made Choosing an AI Even Harder28% of the episode · OpenAI · OpenAI o1 · AppleWatch →AGI by 2030 Is Constrained by More Than Models—It May Run Out of Electricity35% of the episode · Google · Large language model · ChinaWatch →
Open source and open models · 1 episode
Companies in this topic
Topics in this section