How Gemini 3 Is Rewriting Prompting: It’s Not What You Think

How Gemini 3 Is Rewriting Prompting: It’s Not What You Think

Jyunmi opened the show for episode 598 with Andy and Brian, setting up a news heavy Wednesday focused on Gemini 3 and how it changes prompting and agent design. Before diving into Gemini 3, they covered major moves from Nvidia, Microsoft, Anthropic, and Alibaba, plus new tools from Poe and Replit.


Key Points Discussed


Nvidia reports earnings and deepens its partnership with Microsoft and Anthropic, including new chip work tuned for Claude.


Microsoft unveils a sales development agent and an agent command center to track official and shadow agents across 365.


Alibaba launches the Qwen consumer chatbot to compete in China’s crowded assistant market and push deeper ecosystem integration.


Poe adds group chat for up to 200 users with any model, and Replit ships a new design feature powered by Gemini 3.


Google formally launches Gemini 3, wires it into search, the Gemini app, and introduces the anti gravity coding environment.


Brian tests Gemini 3 and finds that it prefers a single large prompt over router style prompt chains.


Gemini 3 introduces an objective based commander intent approach with a prime directive and clear success criteria.


The team walks through new Gemini 3 prompting patterns, including phases instead of steps, deep reasoning loops, and source of truth rules.


Negative constraints and quality gates become core tools to prevent sloppy outputs and premature phase changes.


Brian builds a Gem Architect that helps users design strong Gemini 3 gems using this new prompting style.


He then uses that architect to create a DOS page builder gem that turns show transcripts into SEO ready HTML deep dives.


Andy explains the difference between the Gemini app and Google AI Studio, and how AI Studio is shifting toward full application projects.


Brian shares how the community can access his Gem Architect prompt and gem inside The Daily AI Show hub.


Timestamps & Topics


00:00:00 💡 Intro, episode setup, and agenda

00:01:05 💰 Nvidia earnings and Microsoft Nvidia Anthropic mega deal

00:04:31 🧑‍💼 Microsoft sales development agent and agent command center

00:09:20 🌏 Alibaba’s Qwen consumer chatbot and China price war

00:12:29 🧑‍🤝‍🧑 Poe group chat and Replit design feature with Gemini 3

00:15:34 🤖 Gemini 3 launch, search integration, and anti gravity overview

00:17:13 🧱 From router prompts to mega prompts in Gemini 3

00:22:09 🧭 Objective based commander intent prompting rules

00:26:07 ✅ Negative constraints, quality gates, and phase based flows

00:29:03 🏗️ Gem Architect builder for Gemini 3

00:31:30 📰 DOS page builder gem for Daily AI Show deep dives

00:39:37 🧪 Anti gravity install, hardware notes, and first impressions

00:41:36 🛠️ Finding the AI Studio playground and model options

00:44:53 🧩 Gemini app versus AI Studio and when to use each

00:54:36 🌐 Community hub, prompt sharing plans, and closing


The Daily AI Show Co-Hosts: Andy Halliday, Beth Lyons, Brian Maucere, Jyunmi Hatcher, and Karl Yeh

Det här avsnittet är hämtat från ett öppet RSS-flöde och publiceras inte av Podme. Det kan innehålla reklam.

Avsnitt(892)

Is the AI Slowdown Debate Already Over?

Is the AI Slowdown Debate Already Over?

The hosts discussed responses to Dario Amodei’s call to “pace the frontier,” including opposition from China, President Trump’s rejection of slowing U.S. AI development and NVIDIA CEO Jensen Huang pub...

15 Sep 1h 3min

Can We Slow AI Down Without Losing?

Can We Slow AI Down Without Losing?

The episode centered on a question that suddenly has unusual support across the AI industry: should frontier development slow down enough to give safety systems and institutions time to catch up? The ...

14 Sep 1h 6min

The Watcher-Class Conundrum

The Watcher-Class Conundrum

In OpenAI’s “An Alien Mind,” Jakub Pachocki describes advanced AI as something closer to a grown intellect than a designed machine. Large models emerge from repeated optimization over vast compute, th...

12 Sep 28min

Building An AI First Business -Brian's Demo

Building An AI First Business -Brian's Demo

The episode moved from AI security and platform changes into a live example of what an AI-first business can already look like. Anthropic’s new threat-intelligence report provided the opening story, d...

11 Sep 1h 14min

The Economics of Work In An Age of AI

The Economics of Work In An Age of AI

The episode centered on what happens to the economics of work as AI becomes capable of doing more of it. Anthropic’s new Economic Scenarios Explorer provided the starting point, allowing users to mode...

10 Sep 1h 6min

10,000 AI Agents Attack One Problem

10,000 AI Agents Attack One Problem

The episode opened with the dispute surrounding OpenAI’s newly announced mathematical result and what may be the more important story behind it. Tristan Buckmaster of NYU and Anthropic researcher Leve...

9 Sep 1h 3min

Our Real Atlas Builds and Use Cases

Our Real Atlas Builds and Use Cases

The episode moved quickly from theory to practical experience with GPT-6 Astra. After revisiting OpenAI’s “Alien Mind” paper and the conundrum of using more powerful AI to monitor frontier systems, th...

8 Sep 1h 5min

Can We Truly Control The Alien Mind?

Can We Truly Control The Alien Mind?

The episode focused heavily on GPT-6 Astra and a new essay from OpenAI chief scientist Jakub Pachocki describing advanced AI systems as increasingly alien forms of intelligence that humans grow throug...

7 Sep 54min

Populärt inom Teknik

uppgang-och-fall
market-makers
elbilsveckan
rss-elektrikerpodden
rss-laddstationen-med-elbilen-i-sverige
rss-en-ai-till-kaffet
rss-veckans-ai
skogsforum-podcast
natets-morka-sida
rss-technokratin
rss-ai-med-jonas-benjamin
bli-saker-podden
gubbar-som-tjotar-om-bilar
rss-sakerhetspodcasten
hej-bruksbil
rss-uppgang-och-fall
rss-it-sakerhetspodden
developers-mer-an-bara-kod
rss-nytankarna
rss-digitala-influencer-podden