OpenAI to Produce Its Own AI Chips

Welcome to another edition of AI PlanetX.

OpenAI teams with Broadcom for $10B AI chips; China’s DeepSeek readies autonomous AI agent for late 2025; Atlassian acquires The Browser Company for $610M to create AI browsers.

• Hottest AI News

• Top AI & SaaS Tools

• AI Tutorial: Turn Ideas into UI Designs

• Top AI & Tech News

• AI Art Spotlight

• Prompt of the Day: Life Assessment Prompt

• AI Video Tutorial

• F-R-E-E AI Course of the Day: GPT-5 for Builders by OpenAI

Hottest AI News

OpenAI

OpenAI Partners with Broadcom for $10B Custom AI Chip Production

The AI chip landscape is set for a major shakeup. OpenAI, maker of ChatGPT, is breaking from Nvidia’s grip by developing custom AI processors with Broadcom—signaling a broader shift toward specialized silicon.

Details:

• OpenAI placed a $10B chip order with Broadcom, with production starting next year. It’s the fourth major customer of Broadcom’s custom AI chips for internal use. Broadcom’s stock jumped 9% pre-market and is already up 30% this year

• This puts OpenAI with Google, Amazon, and Meta in building specialized AI chips. CEO Sam Altman flagged compute shortages and plans to double OpenAI’s fleet in five months for GPT-5. The move reduces reliance on Nvidia

• Analysts say this shift could reshape the AI chip market. HSBC predicts Broadcom’s custom unit could surpass Nvidia’s by 2026. While Nvidia still dominates, growth has slowed. The shift to custom silicon shows a maturing industry

This marks a pivotal moment in AI infrastructure. With companies like OpenAI investing billions in custom chips, control over specialized hardware may decide who leads the next phase of AI.

How 1,500+ Marketers Are Using AI to Move Faster in 2025

Is your team using AI like the leaders—or still stuck experimenting?

Masters in Marketing’s AI Trends Report breaks down how top marketers are using tools like ChatGPT, Claude, and Breeze to scale content, personalize outreach, and drive real results.

Inside the report, you’ll discover:

• What AI use cases are delivering the strongest ROI today

• How high-performing teams are integrating AI into workflows

• The biggest blockers slowing others down—and how to avoid them

• A 2025 action plan to upgrade your own AI strategy

Download the report. Free when you subscribe to the Masters in Marketing newsletter.

Learn what’s working now, and what’s next.

DOWNLOAD THE REPORT

High-Flyer

China's DeepSeek Builds Autonomous AI Agent for Late 2025 Launch

China’s DeepSeek is gearing up to shake the AI world again. After stunning the industry with its R1 model, the Hangzhou startup has its sights set on the next big leap: building autonomous AI agents that could rival OpenAI’s latest creations.

Details:

• DeepSeek is building an AI model that can perform multi-step actions with little human input. It aims to handle tasks from travel research to coding while learning from past actions. Founder Liang Wenfeng is targeting a Q4 2025 release

• The project follows DeepSeek's R1, which shook the industry in January by matching OpenAI’s performance at just a few million dollars. Since then, only small upgrades have appeared, with R2 delayed by Liang’s perfectionism

• This push reflects a broader shift toward autonomous AI systems, led by OpenAI, Anthropic, Microsoft, and Manus AI. Unlike aggressive rivals Alibaba and Tencent, DeepSeek prefers a slower, more cautious path

DeepSeek’s upcoming AI agent is a bold move in autonomous intelligence, proving that breakthroughs come from timing and approach, not massive budgets or rushed timelines.

Top AI & SaaS Tools

• Nexuscale (Life-time Deal): Automate B2B lead gen and email outreach with a large leads database, advanced filters, and AI qualification to help sales teams find and convert prospects

• Kimi K2-0905: Mixture-of-experts LLM with stronger coding, 256k-token context window, quicker reasoning, and more accurate tool calls [F-R-E-E]

• CapCut AI: CapCut AI Editing is your smart partner for camera- and prompt-led creation, offering auto edits, avatars, smart cuts, and asset generation [F-R-E-E]

• Incerto: AI-native database copilot that uses context-aware agents to turn natural language into tasks like writing queries, changing schemas, and moving data [F-R-E-E]

• NotebookLM: Latest update turns docs into AI podcasts with adjustable tone, four formats, and flexible durations, making complex info easier to digest [F-R-E-E]

Built for Managers, Not Engineers

AI isn’t just for developers. The AI Report gives business leaders daily, practical insights you can apply to ops, sales, marketing, and strategy.

No tech jargon. No wasted time. Just actionable tools to help you lead smarter.

Start where it counts.

Work smarter daily

AI Tutorial

How to Turn Your App Ideas into Functional UI Designs

Have an app idea but no time to design it from scratch? Let Google’s latest design tool do the heavy lifting! This tutorial will show you how to turn your words into complete UI screens in minutes.

1. Visit Stitch

   Open Stitch in your browser, and choose whether you’re designing for mobile or web.

2. Write app idea

   In the prompt box, describe your app in detail. For example:

   “A fitness tracking app with login, dashboard, activity history, and map integration.”

3. Review plan

   Stitch will create a proposed structure showing all the key screens. Make sure it reflects your idea and the flow you want.

4. Confirm and proceed

   If the plan looks good, tap “Yes” to move forward and let the tool build your design.

5. Wait for UI set

   The AI will automatically generate:

   • Multiple connected screens

   • Layouts and navigation

   • Consistent styling across the app

6. Edit specific screens

   Press “Edit” on any screen to refine details. You can type instructions like:

   • “Move the activities section below the map.”

   • “Add a search bar at the top.”

7. Apply themes

   Switch between dark mode, light mode, or customize fonts and colors to match your brand identity.

8. Preview flow

   Walk through your app’s screens to check navigation and make sure the user experience feels smooth.

9. Export project

   Choose Figma for design refinements or Code to hand off directly to development.

Note: Always preview the full flow of your app before exporting. Checking navigation between screens helps catch usability issues early and ensures the user experience feels smooth.

Top AI & Tech News

• Atlassian is buying The Browser Company for $610M to build AI browsers for knowledge workers, combining its tools with Arc and Dia

• OpenAI plans for an AI hiring platform called OpenAI Jobs Platform to connect businesses with employees and compete with LinkedIn

• Warner Bros. sued Midjourney, claiming it used characters like Superman and Batman without permission in AI-generated images and videos

• Trump hosted top tech CEOs at the White House; Apple, OpenAI, Oracle, SoftBank, and NVIDIA pledged over $1T for U.S. tech investments

AI Art Spotlight

Model: Midjourney V7

Prompt:

A vintage aesthetic scene, an old retro computer monitor standing in a cozy room, on the screen a vivid waterfall is displayed, water flows out of the screen realistically and spills down onto the floor, surrounded by delicate flowers, dreamy retro atmosphere, soft cinematic lighting, slight film grain, photorealistic --ar 9:16 --v 7

Prompt of the Day

This Life Assessment Prompt Terrifies People 😮

A brutally honest life assessment with ten sharp questions that expose where your life is truly headed. No feel-good fluff just a blunt forecast based on your current patterns habits and mindset.

Top AI Video Tutorial

ChatGPT 5 vs. Gemini vs. Claude vs. Grok — The Ultimate Test

This video compares four LLMs—GPT-5 (thinking), Google Gemini Pro, Grok (expert), and Claude Opus 4.1—across 10 tests: UI/visual gen, vision+reasoning, hallucination, how-tos, forecasting, maze gen, spreadsheets, math, info org, and follow-up/context.

Complimentary AI Course of the Day

GPT-5 for Builders

This course teaches you how to use GPT‑5 to ship software more efficiently and with higher quality. You’ll learn practical workflows for refactoring large codebases, writing and reviewing code, integrating unconstrained function calling, designing structured outputs, and orchestrating multi‑tool pipelines.