God Mode Podcast

God Mode Pod

Every week, Rik, Ben & Luca break down the biggest AI stories. The drama, the launches, the "who just got disrupted" moments, through three lenses: marketing, building, and macro vision. For solopreneurs, indie hackers, AI enthusiasts, and idea people becoming founders. New episode every Saturday.

  1. 3d ago

    EP16: Nobody's seen AI UGC work, world models & AI agents go desktop

    Four months in, all three of us accidentally wore white shirts — then spent an hour arguing about whether AI actually works. Icon.com is back: the founder sold Skio for $105M, spent $12M on the domain, hit $5M ARR in 30 days, went dark — and relaunched by walking back "AI ad maker" to "human ad maker." We use it to ask the real question: has anyone actually seen AI UGC work? Then a full media-buying playbook ($1 vs $30 CPM, Facebook pixel seasoning), Fei-Fei Li's world models (what comes after the LLM), the AI bubble and the fiber analogy, Nous Research's Hermes desktop AI agent, a blind game guessing 5 AI coding IDEs that all look the same, and AI-slop design. In this episode: • Icon.com's relaunch and the AI-UGC walk-back • Why human UGC still beats AI UGC — and the $1 vs $30 CPM math • Facebook pixel, account seasoning & being TikTok-famous to lower ad costs • Fei-Fei Li's world models: renderers, simulators, planners • Opus 4.8 does 80% of desk work — robots are the hard part • Is the AI bubble about to pop? The fiber overinvestment analogy • Nous Research's Hermes desktop app & the IDEs that all look the same • AI-slop design and writing a book trilogy with AI Chapters: 00:00 Cold open — the accidental white-shirt gang 01:34 "DNA is just code" — Luca's pharma → tech crossover 02:53 Icon.com is back — the $12M domain comeback 07:00 Has anyone actually seen AI UGC work? 15:44 UGC economics — $1 vs $30 CPM 21:18 Facebook pixel, seasoning accounts & Meta's AI targeting 23:52 Fei-Fei Li's world models — what comes after the LLM 28:56 AI philosophers: Machiavelli vs Marcus Aurelius 31:07 Open-sourced voice, 110ms latency & the freemium model 35:43 Is the AI bubble about to pop? Uber, IPOs & the fiber analogy 38:53 Nous Research Hermes — agents move to the desktop 44:59 The IDE game — "they all look the same dude" 47:23 AI slop design & the "Signs of AI Writing" checklist 51:52 Writing books with AI — Luca's trilogy 54:28 Outro — see you in Malta Hosts: Rik — x.com/rikventure Ben Broch — x.com/brochchain Luca Arrigo — x.com/LucaArrigo_ Watch on YouTube → https://www.youtube.com/channel/UCUhKe2X1RW-kLTDigvkMz4w Newsletter → https://godmodepod.substack.com/ Follow the show for a new episode every Saturday.

    53 min

  2. May 29

    Claude Opus 4.8, browser in Codex, tokenmaxxing, and the upcoming AI IPO bear case nobody wants to hear

    Claude Opus 4.8 just dropped, Anthropic is renting SpaceX GPUs at $8/hour, and SpaceX filed toward the biggest IPO in history. Plus token maxing, agentic onboarding, and analyzing your own DNA with AI. Episode 15 of God Mode Pod with Rik, Ben and Luca. We hit one million views — then spent the week questioning everything: is Opus 4.8 a real release or PR for the next raise, why "token maxing" is becoming a vanity metric, which model you should actually run for which task, and how agentic onboarding changes how every app ships. Then a full SpaceX IPO breakdown: the $8/hour Anthropic GPU deal, a TAM bigger than China, the East India Company parallel, and why the bear case might be stronger than the Elon FOMO. We close with Blue Origin going boom, Content of the Week (YC Paper Club, Dan Shipper, Lenny's pod), and a Science Corner on analyzing your own DNA with AI. Nothing here is financial or medical advice. CHAPTERS 00:00 We hit 1 million views 00:35 Intro — Luca's 24 hours in Vienna & the Malta meetup 02:30 Claude Opus 4.8: real release or PR for the raise? 06:00 Token maxing: burn everything vs return-on-token-spend 14:30 Which model for which task — the subscription stack 22:00 The browser inside Codex & agentic onboarding 30:00 Have GPU prices peaked? The $8/hr Anthropic–SpaceX math 36:00 SpaceX IPO deep dive — the bear case 54:00 Quickfire: Pieter Levels vs Google Maps & the $500M Amazon bill 1:00:00 Blue Origin goes boom 1:02:00 Content of the Week — YC Paper Club, Dan Shipper, Lenny's pod 1:08:00 Science Corner — analyzing your own DNA with AI 1:16:00 Outro HOSTS Rik — x.com/rikventure Ben Broch — x.com/brochchain Luca Arrigo — x.com/LucaArrigo_ Watch on YouTube: https://www.youtube.com/@godmodepod Newsletter: https://godmodepod.substack.com/ New episodes every Saturday.

    1h 27m

  3. May 22

    SpaceX's $2T IPO, Karpathy Joins Anthropic & the SaaSpocalypse — ft. Dr Dan | EP14

    SpaceX filed the biggest IPO in history. Andrej Karpathy left OpenAI for Anthropic. And the AI coding wars got brutal: Cursor vs Codex vs Claude Code, the SaaSpocalypse, plus a full Ozempic breakdown. Episode 14 with guest Dr Dan (ex-Google, now a master vibe coder building OmniHarness.dev). Rik, Ben and Dan go deep on the $1.75 trillion SpaceX IPO and the plan to acquire Cursor, why Karpathy joining Anthropic flipped the script, whether B2B SaaS is already a one-click clone, and the real economics behind every AI coding subscription. Then a full breakdown of retatrutide, semaglutide and the peptide wave: why Ozempic is $20 from China and $400 everywhere else. Nothing here is financial or medical advice. CHAPTERS 00:00 Cold open + who's on the show (Dr Dan, ex-Google) 02:50 SpaceX files the biggest IPO in history ($1.75T) 05:00 Data centers in space & Elon's master narrative 09:45 SpaceX wants Cursor: does Cursor have a moat? 14:00 Codex built an app in 7 hours while he slept 17:00 He fired 18 people & cloned a $34k SaaS from a call 19:00 Does Google copy your app? (Theo) 20:00 OpenAI's $2M-per-YC-startup playbook 25:30 Memory is the real AI lock-in 27:30 Karpathy joins Anthropic: is it "all hype"? 31:00 You don't need frontier models (and the subsidy game) 38:30 Meta layoffs & the fast-fashion software era 40:00 The SaaSpocalypse: every company becomes a software company 43:30 The counter: mom-and-pops won't vibe code 48:00 Inside China: a "side hustle" is a 40-person factory 51:30 Retatrutide & the Pokémon evolution of Ozempic 1:00:30 Why Ozempic is $20 from China (the loophole) 1:02:00 Are peptides just placebo? + the washing-machine take 1:05:30 Outro HOSTS Rik — x.com/rikventure Ben Broch — x.com/brochchain Guest: Dr Dan — OmniHarness.dev Watch on YouTube: https://www.youtube.com/@godmodepod Newsletter: https://godmodepod.substack.com/ New episodes every Saturday.

    1h 7m

  4. May 16

    Anthropic won the AI race, OpenAI gives Codex for free, xAI Grok Build & Claude cancel culture | GMP EP13

    Anthropic beat OpenAI on business adoption, xAI launched Grok Build, and Rik cancelled his $100 Claude plan on air. The AI news that actually matters for vibe coders, AI builders and indie hackers — God Mode Pod EP13, with Rik, Ben and guest Matt. In this episode: - Why Rik cancelled his $100 Claude Max plan and split the spend four ways - Ben reframes the viral "AI bubble" chart — one coder burns 100x the tokens of a casual ChatGPT user - Anthropic beats OpenAI on business adoption for the first time - The /goals feature: Codex sold 200 items in 4 days from a single instruction - HTML is the new Markdown — Rik's sold, Matt's skeptical - Anthropic's June 15 credit change and the $7,500-vs-$200 lock-in - Grok Build: xAI finally enters the coding-agent race - Matt's "don't check" tip that speeds up Codex and Claude Code Chapters: 00:00 Cold open 00:35 Three vibe coders, one podcast (Matt joins) 01:30 Is AI video editing finally close? 03:50 The AI super cycle + the conglomerate called Elon 08:00 Inside Amazon's robot warehouses 09:30 The AI echo chamber: the one red dot 14:00 AI therapists, the sleeper use case 18:30 Anthropic beats OpenAI on business adoption 21:00 Cursor vs Claude Code: the harness wars 29:35 Why Rik cancelled his $100 Claude plan 33:30 HTML is the new Markdown 40:30 /goals: agents that don't stop 49:30 Anthropic's credit clampdown + the monopoly question 55:30 Grok Build: xAI enters the ring 01:02:00 Voice cloning gets weird 01:04:30 Notion opens up + Peter Levels' cheap stack 01:19:30 Who builds the AI video editor? 01:25:30 Google Book, space data centers & Clicky 01:33:30 The China trip + Elon meme mode 01:37:00 Outro God Mode Pod — three builders breaking down the AI news that matters, every Saturday. Watch on YouTube: https://youtube.com/@godmodepod Newsletter: https://godmodepod.substack.com/ Hosts: Rik — x.com/rikventure Ben Broch — x.com/brochchain Guest: Matt -- x.com/MattPenny99

    1h 38m
  5. Anthropic + SpaceX, Codex 100x'd Claude Code & Half Cash Half Stock | GMP EP12

    May 8

    Anthropic + SpaceX, Codex 100x'd Claude Code & Half Cash Half Stock | GMP EP12

    Anthropic just shook hands with SpaceX. Codex now has 100x more downloads than Claude Code. Anthropic stacked $380 billion in compute and 10% of America's data center capacity in one week. The agent layer is winning. Codex 46M downloads vs Claude Code 491K. Anthropic + SpaceX partnership doubles rate limits with multi-gigawatt orbital compute on the roadmap. Code with Claude 2026 ships Dreaming + multi-agent orchestration — every announcement was platform, not model. Friend of the pod Alan from the Philippines joins to break down Hermes vs OpenClaw (Honda vs Ferrari, credit Gary Tan). Luca calls PrintingPress.dev "the new WordPress." Apple iOS 27 picks Claude + Gemini, OpenAI nowhere on the list. Google Fitbit Air drops at $99 one-time vs Whoop's $260/year. Neuralink hits 21 people typing 40 wpm with their thoughts. And the closer: Ryan Cohen's "half cash, half stock" CNBC interview — the most engineered viral moment in business of 2026. Three builders. Weekly AI news. No fluff. The big thesis: compute won the lab war. Agents are about to win the app war. If you're building anything in 2026, the question isn't "is SaaS dead" — it's who builds the Hermes for everyone else? 🎙️ Hosts: Rik — x.com/rikventure Ben Broch — x.com/brochchain Luca Arrigo — x.com/LucaArrigo_ Watch on YouTube → https://youtube.com/@godmodepod Substack → https://godmodepod.substack.com/

    1h 55m
  6. Codex beats Claude Code, Stripe launched the AI bank & SpaceX targets a $7.5T valuation | GMP EP11

    May 1

    Codex beats Claude Code, Stripe launched the AI bank & SpaceX targets a $7.5T valuation | GMP EP11

    This was the week Cursor beat Anthropic at running Anthropic's own model. The third-party harness extracts more performance from Opus 4.7 than Claude Code does. Codex ate Claude Code with HeyGen and browser baked in. Grok 4.3 dropped at one-quarter OpenAI's pricing. GitHub Copilot moved to usage-based billing. Anthropic shipped Claude Security and Cursor cloned it in two hours flat. Stripe Treasury launched with Stripe MCP — the financial wallet for AI agents. SpaceX board approved Elon's $7.5T pay package with a 100 terawatt compute clause and a 1-million-person Mars colony clause. And the Department of War named its seven AI partners on Friday morning — Anthropic was notably excluded. Three builders. Weekly AI news. No fluff. The big thesis: the abstraction layer is eating the lab. Crypto already ran this movie — Phantom prints, Solana doesn't. AI is the remake. 🎙️ Hosts: Rik — x.com/rikventure Ben Broch — x.com/brochchain Luca Arrigo — x.com/LucaArrigo_ Watch on YouTube → https://youtube.com/@godmodepod Substack → https://godmodepod.substack.com/

    1h 25m
  7. Anthropic apologized, SpaceX wants to buy Cursor & hairdryer hacked Polymarket | GMP EP10

    Apr 25

    Anthropic apologized, SpaceX wants to buy Cursor & hairdryer hacked Polymarket | GMP EP10

    This was the wildest seven days of AI news in 2026 — and we covered all of it. Anthropic finally admitted what builders had been saying for six weeks: Claude got dumber. They confirmed three product-layer bugs stacked on top of each other from March 4th to April 16th. Default reasoning silently dropped from "high" to "medium." A caching bug killed idle session thinking. A system prompt told Claude to keep responses under 100 words, truncating reasoning chains across the entire product. None of it was a model change. All of it was scaffolding. Then GPT 5.5 dropped — better benchmarks than Opus 4.7, faster, co-designed with NVIDIA on 100,000 Blackwell GPUs. But the number nobody put on the slide: 86% hallucination rate. Up from 52% on Opus 4.7. Way up from Grok at 17%. The benchmark race is the wrong race. Plus: SpaceX got a $60B call option to buy Cursor — the world's largest call option, structured as a talent grab to replace Elon's burned XAI founding team. Naval launched USVC, opening late-stage AI venture exposure to retail at $500. Ben breaks down the 30:1 valuation pattern showing up across the entire AI sector — Scale AI, Cursor, Anthropic, every single one priced at the same multiple. And a guy made $34,000 betting on Paris weather by driving to the weather station with a hairdryer. Three builders. Weekly AI news. No fluff. ⏱️ CHAPTERS 00:00 Cold open — gateway drug + 86% hallucination + compute thesis 02:30 Anthropic apologizes — three stacked Claude bugs explained 11:30 Spoon-feeding theory + Claude as gateway drug 19:00 90% of my week in Claude desktop — Anthropic's everything app 20:45 Provider or software? Anthropic must pick 23:00 No world where compute goes unused — Ben's bubble counter 26:42 Coding is a circular reference — the AGI flywheel 29:00 Domino theory of the AI bubble 32:11 SpaceX $60B Cursor deal — talent grab, not software 32:28 The 30:1 rule of AI valuations 41:53 GPT 5.5 hallucinates 86% 45:24 OpenAI exec exodus — $500M unvested equity 52:31 Is Claude becoming a verb? 58:00 Naval drops USVC — retail venture for $500 1:08:24 Xpeng flying cars 7,000 pre-orders 1:11:00 Polymarket gets hacked with a hairdryer 1:11:13 Luca's 2040 mountain hut vibe-code future 1:15:40 Polymarket isn't trustless 🎙️ HOSTS Rik — x.com/rikventure Ben Broch — x.com/brochchain Luca Arrigo — x.com/LucaArrigo_ Listen on Spotify → https://open.spotify.com/show/0ZBhl5IVFa8wZPvIKNTkRP Apple Podcasts → https://podcasts.apple.com/us/podcast/god-mode-podcast/id1883959580 Substack → https://godmodepod.substack.com/

    1h 18m
  8. Anthropic made Opus dumber (73% Drop), Claude Code Hits $2.5B ARR & Tesla AI5 | God Mode Pod EP09

    Apr 18

    Anthropic made Opus dumber (73% Drop), Claude Code Hits $2.5B ARR & Tesla AI5 | God Mode Pod EP09

    Anthropic pre-IPO hit $900B. Claude Code crossed $2.5B ARR in nine months — the fastest B2B growth ever recorded. And your Claude Opus didn't get dumber — Anthropic quietly nerfed it (73% drop in reasoning depth across 7,000 sessions). This week on God Mode Pod: Rik, Ben, and Luca break down the three-week Anthropic news dump — the Tesla vs NVIDIA playbook debate, the Opus reasoning rollback, the desktop Claude Code redesign, and whether IDEs will actually be dead by end of year. Plus: Tesla AI5 tapes out (2,500 TOPS, 17x jump from 2019). The OpenAI CRO memo leak. Luca's antitrust prediction. A 24-year-old ex-OpenAI researcher who turned a $254M fund into $5.5B in one year. Allbirds pivoting to AI compute for a 910% single-session run. And the best builder story of the week — a pool postcard AI agent that's already making money. Three builders. One hour. Everything that mattered in AI this week. New episode every Saturday.

    1h 9m
See All (16)

About

Every week, Rik, Ben & Luca break down the biggest AI stories. The drama, the launches, the "who just got disrupted" moments, through three lenses: marketing, building, and macro vision. For solopreneurs, indie hackers, AI enthusiasts, and idea people becoming founders. New episode every Saturday.

Information

  • Creator
    God Mode Pod
  • Years Active
    2K
  • Episodes
    16
  • Rating
    Explicit
  • Copyright
    © God Mode Pod

You Might Also Like