TNB Tech Minute: OpenAI And Broadcom Develop a Custom Chip for AI Inference

Summary

WSJ Tech News Briefing – TNB Tech Minute: OpenAI And Broadcom Develop a Custom Chip for AI Inference

Date: June 24, 2026
Host: The Wall Street Journal (Julie Chang)
Episode Focus: Latest developments in AI chip manufacturing and software, major tech M&A, and robo-taxi innovation.

---

Episode Overview

This fast-paced Tech Minute highlights key advancements in the AI hardware and software landscape. The spotlight is on a significant collaboration between OpenAI and Broadcom to develop a custom AI inference chip, major industry moves like Qualcomm’s acquisition of Modular, and a product update from Amazon’s robo-taxi subsidiary, Zoox. The episode also includes expert commentary on the business challenges of scaling AI within organizations.

---

Key Discussion Points & Insights

1. OpenAI & Broadcom Unveil Custom AI Inference Chip

• Main News:
  • OpenAI joins forces with Broadcom to create "Jalapeno," a custom chip built specifically for efficient, large-scale AI model inference.
  • Julie Chang (00:26):

    "The chip, dubbed Jalapeno, was built from scratch to improve efficiency and lower costs for serving AI systems at scale."

• Details:
  • Neither company provided specific performance metrics, but a detailed technical report is promised in the coming months (00:41).
  • The initiative is core to OpenAI’s strategy to lessen dependency on outside hardware suppliers and to tailor infrastructure for increasing AI deployments (00:48).
  • Broadcom frames the project as the first step in a multi-generation partnership targeting large-scale AI infrastructure.

2. Qualcomm to Acquire Modular in $3.9 Billion Deal

• Announcement:
  • Qualcomm is set to buy Modular, an AI software company, in a $3.9 billion stock transaction (01:07).
  • Purpose:
    • A move to make AI more accessible, faster, and economical for Qualcomm’s customers.
  • Julie Chang (01:14):

    "AI faster and cheaper for the semiconductor company's customers."

• About Modular:
  • Produces software designed to let AI applications run smoothly across diverse hardware platforms, making deployment cheaper and easier for developers (01:20).
  • The deal is anticipated to close in the second half of 2026, with an aim to enable efficient scaling of AI from device to cloud for clients (01:29).

3. Zoox’s Redesigned Robo-Taxi Debuts

• Product Launch:
  • Zoox, an Amazon subsidiary, introduces an updated robo-taxi model optimized for mass production, while retaining its signature carriage-style cabin (01:46).
• Key Features & Updates:
  • Lighter interior and refined exterior for enhanced aesthetic and functional distinction (front vs. back).
  • Incorporation of a fresh speaker and microphone in the door interface for rider interaction (01:47).
• Production and Rollout:
  • Zoox will begin large-scale manufacturing at its Hayward, California plant, with plans to produce 100 vehicles per week.
  • Availability to riders is expected later in 2026, pending regulatory approval (02:00).

4. Expert Insight: Scaling AI Beyond Technology

• Guest Expert: Caroline Roach, Senior Partner, IBM Consulting
• Key Challenge:
  • Many organizations stall their AI projects not due to tech limits, but because of organizational “misalignment”—unclear goals, suboptimal processes, and misaligned incentives (00:00 and returns at 02:19).
  • Caroline Roach (02:25):

    "The biggest thing that we were talking about a year ago is what model to use, and the biggest thing that I'm talking about with my clients now is how do I drive change within my organization?"

• Success Factors:
  • Organizations that set clear targets and company-wide priorities are positioned to realize the true business value of AI.
  • Real implementation value is rooted in organizational change, updated incentives, and redesigned workflows—not tech alone.
  • Caroline Roach (02:44):

    "The technology is really good, but if you're not changing your organizational alignment, not incentivizing your people correctly, not looking at workflows—you're not going to see real value with it."

---

Memorable Quotes

• Julie Chang (00:26):

  "The chip, dubbed Jalapeno, was built from scratch to improve efficiency and lower costs for serving AI systems at scale."

• Julie Chang (01:14):

  "AI faster and cheaper for the semiconductor company's customers."

• Caroline Roach (02:25):

  "The biggest thing that we were talking about a year ago is what model to use, and the biggest thing that I'm talking about with my clients now is how do I drive change within my organization?"

• Caroline Roach (02:44):

  "The technology is really good, but if you're not changing your organizational alignment, not incentivizing your people correctly, not looking at workflows—you're not going to see real value with it."

---

Timestamps for Major Segments

• 00:26 – OpenAI & Broadcom announce Jalapeno AI chip
• 01:07 – Qualcomm to acquire Modular for $3.9 billion
• 01:46 – Zoox debuts new carriage-style robo-taxi, eyes large-scale rollout
• 02:19 – Caroline Roach on the organizational side of scaling AI

---

Episode Highlights

• The AI industry is moving beyond off-the-shelf hardware, with custom chips such as Jalapeno designed for ever-larger, more demanding language models.
• M&A activity in AI software continues as hardware vendors seek tighter integration for efficient deployments.
• The robo-taxi market is advancing toward mass production, with Zoox’s design updates aimed at practical rollout.
• Success in scaling AI remains as much about organizational change as it is about technological progress—clear priorities, incentives, and processes are key.

---

For listeners wanting the latest on AI’s technological and business evolution, this episode provides essential updates and actionable insights in a quick, engaging format.

Transcript

[Caroline Roach]

Many companies are struggling to scale their AI deployments or even move them past the pilot stage. Often the problem isn't technology, but organizational misalignment around goals, processes and incentives. At the break, join Caroline Roach, Senior Partner, IBM Consulting, to learn why.

[TNB Tech Minute Host]

Here's your afternoon TNB Tech minute for Wednesday, June 24th.

[Julie Chang]

I'm Julie Chang for the Wall Street Journal.

[TNB Tech Minute Host]

OpenAI and Broadcom have joined forces to

[Julie Chang]

develop a custom AI chip for large language model inference. The chip, dubbed Jalapeno, was built from scratch to improve efficiency and lower costs for serving AI systems at scale.

[TNB Tech Minute Host]

OpenAI and Broadcom didn't disclose specific performance

[Julie Chang]

measures, but said a more detailed report will be available in the coming months.

[TNB Tech Minute Host]

The project is part of OpenAI's strategy

[Julie Chang]

to reduce reliance on external hardware suppliers and optimize infrastructure for its growing AI services. Broadcom said the project represents the start of a multi generation collaboration aimed at building large scale AI infrastructure.

[TNB Tech Minute Host]

Qualcomm has agreed to acquire AI software

[Julie Chang]

company modular in a $3.9 billion stock deal.

[TNB Tech Minute Host]

The acquisition is an effort to make

[Julie Chang]

AI faster and cheaper for the semiconductor company's customers.

[TNB Tech Minute Host]

Modular makes software that allows AI to

[Julie Chang]

run efficiently across hardware architectures, enabling developers to deploy AI at a lower cost.

[TNB Tech Minute Host]

Qualcomm said it aims to help customers

[Julie Chang]

move AI into production from device to cloud more quickly. The deal expected to close in the second half of this year, and Amazon subsidiary Zoox debuted a redesigned Robo taxi today optimized for mass production while retaining

[TNB Tech Minute Host]

its carriage style cabin.

[Julie Chang]

The update includes a lighter interior palette, refined exterior to differentiate the vehicle's front from the rear, and a new speaker and microphone in the door interface.

[TNB Tech Minute Host]

Zoox plans to begin large scale manufacturing

[Julie Chang]

at its Hayward, California plant sales soon and expects it'll become available to riders later this year pending regulatory approval. The company said production could reach 100 vehicles weekly to facilitate this year's expansion.

[TNB Tech Minute Host]

And that's your TMB tech minutes.

[Julie Chang]

Check back in the morning for another quick tech update.

[Caroline Roach]

Scaling AI successfully requires more than the right technology. Here again is Caroline Roach, Senior partner,

IBM Consulting the biggest thing that we were talking about a year ago is what model to use and the biggest thing that I'm talking about with my clients now is how do I drive change within my organization?

Companies able to identify correct and then avoid misalignment will be best positioned to deliver meaningful business value from AI.

The organizations that are the most successful set very clear targets and have several priorities that are very clear across the enterprise. The technology is really good, but if you're not changing your organizational alignment, not incentivizing your people correctly, not looking at workflows. You're not going to see real value with it.

To learn how building organizational alignment can help deliver AI deployments that scale and drive growth.

[WSJ Advertising Department]

This content was created by custom content from WSJ, a unit of the Wall Street Journal advertising department.