Join us on the STILL RELEVANT tour: https://simulationtheory.ai/16c0d1db-a8d0-4ac9-bae3-d25074589a80
Join Simtheory: https://simtheory.ai

🚀 Try our AI-built apps:

Macrosoft Teams: teams.simtheoryapp.com (working video chat with up to 150 people)
Trallo: trallo.simtheoryapp.com (full Trello clone, unlimited boards, completely free)

TDIA Discord: https://discord.gg/gTW4RkAJvn
Spotify Songs: https://open.spotify.com/artist/28PU4ypB18QZTotml8tMDq?si=Zh4jgHIASI2ZvsXVfVcCoA

So Chris, this week... we've been having way too much fun with the AI again. OpenAI just dropped GPT-5.4 and 5.4 Pro, and holy shit - we finally have a ball game. This might be the first OpenAI model that genuinely competes with Opus 4.6 for agentic work.

But here's where it gets wild: we rebuilt Trello AND Microsoft Teams from scratch using single prompts. Not mockups. Fully deployed, working apps with authentication, video chat, the works. You can literally sign up and use them right now.

Plus: We roast Gemini 3.1 (it's a disgrace for agentic workflows), break down the insane $30/$180 per million pricing on 5.4 Pro (who is this for??), and discuss why every $99/month SaaS tool might be about to die. Chris declares his programming skills "useless" and honestly... he might be right.

We also demo our actual workflow - running 5 agent tabs simultaneously, delegating everything, and why we barely visit websites anymore. The AI workspace IS the operating system now.

CHAPTERS:

0:00 - Intro & Housekeeping (We Screwed Up the Link)
1:26 - GPT-5.4 First Impressions & Specs
3:12 - Chris's Testing: 40 Minutes to Solve a Problem
4:51 - Knowledge Work Improvements (Catching Up to Anthropic)
6:38 - Computer Use vs Browser/Terminal Debate
8:07 - Why We Don't Need Computer Use Anymore
9:53 - Teaser: We Built Full SaaS Apps Today
11:19 - Tool Search API & Skills Integration
13:20 - The Speed Problem (It's a Plodder)
15:12 - GPT-5.4 Pro Pricing Reaction ($30/$180 WTF)
18:14 - Someone Rebuilt Minecraft in 24 Minutes
19:46 - Gemini 3.1 Roast: "It's a Disgrace"
22:36 - DEMO: Trallo (Full Trello Clone)
29:03 - DEMO: Macrosoft Teams (Working Video Chat!)
33:30 - The SaaS Collapse Theory
36:42 - AI Workspace as the New Operating System
38:57 - Forcing Features onto Entrenched Software
43:32 - "My Programming Skills Are Useless" - Chris
46:06 - The $12 Million Legacy Software Opportunity
51:06 - Beyond Code: Forms, PDFs, Knowledge Work
55:28 - How Fast Will This Change Everything?
56:31 - Gemini 3.1 Flash Lite Quick Take
59:36 - The Delegation Lifestyle (5 Agent Tabs Running)
1:01:24 - Mike's Workflow Demo
1:04:31 - Cognitive Overload Problem
1:06:04 - Release Date: 2 Weeks (Drop Punishment Ideas!)

Thanks for listening like and sub xoxo

Join us on the STILL RELEVANT tour: https://simulationtheory.ai/16c0d1db-a8d0-4ac9-bae3-d25074589a80

Join Simtheory: https://simtheory.ai

TDIA Discord: https://discord.gg/gTW4RkAJvn

Horse Egg Lifecycle Infographic: https://staging.simtheory.ai/share/file/UZ2KJU

----

So Chris, this week... we're diving into Google's new Nano Banana 2 image model - 50% cheaper and supposedly faster (when the servers aren't melting). We put it through its paces with annotation-based editing, slide generation, and yes, the return of the legendary horse egg experiment.

Plus: Google quietly kills Gemini-3 after just a few months (good riddance?), we discuss why the model was "dead on arrival" for agentic workflows, and break down the real story behind those massive AI layoff announcements from Block and WiseTech. Spoiler: it's probably not actually about AI.

We also get into the current state of the model wars (Opus 4.6 vs Codex 5.3), why smaller models like GLM-5 might be the future for enterprise agentic tasks, and Chris's wife teaching Claude to literally speak to her using Mac's text-to-speech. The models are getting creative.

---
0:00 - Intro
0:36 - Nano Banana 2: Price, Speed & First Impressions
3:19 - The Compositing Problem & Last Mile Design
5:41 - Annotation-Based Editing (This Changes Everything)
9:52 - Slide Editing & Real-World Use Cases
12:34 - The Horse Egg Experiment Returns
14:30 - Image Degradation & Cost Breakdown
17:47 - Text-to-Image Leaderboard Discussion
20:01 - Why Nano Banana Dominates for Work
22:07 - Codex 5.3 vs Opus 4.6
22:54 - Google Kills Gemini-3 (What Went Wrong?)
26:48 - Google's Agentic Problem
30:08 - The Model Loyalty Cycle
34:22 - Why Opus 4.6 is Still the Best
37:05 - Cost Optimization & Smart Model Routing
43:30 - When Models Get Stuck on the Wrong Path
45:36 - Nicole's AI Learns to Talk Back
46:54 - Can Anyone Build Software Now?
52:26 - Anthropic's Legal/Finance Plugins & Market Panic
57:08 - Block Lays Off 4,000: AI or Excuse?
1:00:05 - The AI Job Apocalypse Isn't Real

Thanks for listening like and sub xoxo

• Join Simtheory: https://simtheory.ai
• "Is This The End" now on Spotify: https://open.spotify.com/album/2Py1MyADUFqJFVUISI2VTP?si=oT3PWyJYRA2BspOmzT_ifg
• Register for the STILL RELEVANT tour: https://simulationtheory.ai/16c0dationtheory.ai/16c0d1db-a8d0-4ac9-bae3-d25074589a80

Two new models dropped this week — Gemini 3.1 Pro and Claude Sonnet 4.6 — and honestly? We're struggling to care. In this episode, we break down why Gemini went from being our daily driver to a model we barely touch, the "tunnel vision" hallucination problem that killed the Gemini 3 series for us, and whether 3.1 Pro actually fixes it. We put Gemini 3.1 Pro head-to-head against Claude Opus building a Geoffrey Hinton Doom Center, debate whether anyone can actually tell the difference between Sonnet 4.5 and 4.6, and make the case that smaller models running in agentic loops are secretly beating the frontiers. Plus: OpenAI acquires OpenClaw and we ask why a $100B company couldn't just build it themselves, DHH calls out the AI pricing bubble, Mike compares AI models to cheap wine hangovers, and Sam Altman refuses to hold Dario's hand at the India AI Summit. The model wars are getting weird.

CHAPTERS:

0:00 Intro & "Is This The End" Now on Spotify
1:10 Gemini 3.1 Pro: Thinking Controls & The Medium Mode Fix
3:14 The Speed vs Intelligence Trade-Off in Agentic Work
5:10 Why Multitasking With AI Agents Made Us Anxious
6:34 Solid Updates: The Real Goal of Agentic Coding
7:45 Gemini's Fall From Grace: From Daily Driver to Dead Model
10:08 The Tunnel Vision Problem That Killed Gemini 3
13:35 Mixed Reactions: Fanboys vs Reality on Gemini 3.1 Pro
15:06 Side-by-Side Test: Gemini 3.1 Pro vs Claude Opus (Hinton Doom Center)
17:39 Why File Manipulation Accuracy Matters More Than Context Windows
19:27 The Context Window Debate: 1M Tokens vs Smart Sub-Agents
22:05 DHH on Token Pricing: "If There's a Bubble, It's This"
24:11 Should Models Ship as Agent vs Chat Variants?
28:43 Claude Sonnet 4.6: A $2 Discount on Opus?
31:44 The Model Mix: Why One Model Won't Rule Them All
34:40 Anthropic Is Winning — But Can Anyone Tell the Difference?
38:58 OpenAI Acquires OpenClaw: Why Couldn't They Just Build It?
44:18 The Silicon Valley Moment: Sam vs Dario at India AI Summit
47:05 Will Smaller Models Win the Enterprise? The Cost Reality Check
51:27 The End of Single-Shot: Why Agentic Loops Change Everything
55:48 Final Thoughts & Gemini 3.1 Pro Gets One More Week

Thanks for listening. Like & Sub. Links above for the Still Relevant Tour signup and Simtheory. Two models dropped on a week again. What a time to be alive. xoxo