The Information
Broadcom Reports Big AI Chip Revenue Increase -- China Considers Record $70 Billion Chip Incentive Program -- OpenAI Releases New Flagship AI Model -- Nvidia To Host Summit On Data Center Power Shortage  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ 

Now streaming → →

Dec 12, 2025

The Information AM

Supported by Sponsor Logo

Save 25% on an annual subscription to read the most important news about technology and business first. For even more, Save $250 on The Information Pro for unlimited access to our proprietary org charts, databases and surveys.

Thank goodness it's Friday! Disney will invest $1 billion in OpenAI and license 200 of its best known characters for video generation in the Sora app. Broadcom reports a big jump in its AI chip revenue. China considers a record $70 billion incentive program to support its chip sector.

Read more briefings
1.
Disney to Invest $1 Billion in OpenAI
By Martin Peers Source: The Information

Walt Disney Co. is investing $1 billion into OpenAI as part of a deal in which Disney is licensing more than 200 characters from its program library for use in OpenAI’s Sora video-generating app. Among the characters that people will be able to use in Sora videos are Mickey Mouse, Cinderella, Black Panther, Captain America, Iron Man, Darth Vader and Luke Skywalker.

The deal follows criticism from Hollywood figures that Sora could be used to steal images of actors without compensation. In striking a deal with Disney, OpenAI demonstrates it is willing to work with Hollywood. Disney drew an implicit contrast with Google: on Wednesday, Disney lawyers sent Google a letter demanding it stop infringing on its copyrights in its training of AI models, Variety reported.

The announcement from OpenAI and Disney didn’t make clear, however, what the terms of the licensing agreement are, and whether OpenAI is paying money. Instead, what is clear is that Disney will be paying OpenAI. In addition to investing in OpenAI, Disney will become a major customer of OpenAI, the companies said, using the AI firm’s tools to build new products and experiences including for its Disney+ streaming service.

Disney will be able to run some of the Sora videos on its streaming service, Disney+. The videos will include “costumes, props, vehicles and iconic environments” from the programs but not “talent likenesses or voices.”

Disney CEO Bob Iger has ties to OpenAI, through Thrive Capital, one of OpenAI’s biggest backers. Iger was briefly a venture partner at Thrive and took a stake in its management company in 2023.  He remains a limited partner in its funds.

2.
Broadcom Reports Big AI Chip Revenue Increase
By Martin Peers Source: The Information

Broadcom reported 28% higher revenue of $18 billion for its fourth fiscal quarter ended Nov. 2, driven by a 74% jump in AI chip revenue, reflecting how its business of helping companies like Google develop specialized AI chips is lifting Broadcom’s business significantly. Broadcom shares rose 3% in -after-hours trading.

But AI chips are less than half of Broadcom’s business and other parts of the company are performing less robustly, which dilutes the growth impact. Broadcom CEO Hock Tan forecast that AI chip revenue would double in the company’s first fiscal quarter, through January, while total revenue would only rise 28% to $19.1 billion, the same growth rate as the fourth quarter.

Broadcom’s free cash flow is growing faster than its topline, however. In the fourth quarter, free cash flow hit $7.466 billion, up 36% on the year-earlier quarter.

Tan also said on the firm’s earnings call that Anthropic was the customer who placed a $10 billion order last quarter. He said Anthropic bought racks of Google’s tensor processing units. Anthropic placed another $11 billion order in the current quarter for the chips, Tan added. The Information first reported on Google’s plans to sell its TPUs directly to customers, instead of renting them through Google Cloud, in an effort to compete more directly with Nvidia.

3.
China Considers Record $70 Billion Chip Incentive Program
By Qianer Liu Source: Bloomberg

Beijing is considering pouring as much as $70 billion into its semiconductor industry in what would be the world’s largest state-backed chip support program, Bloomberg reported, citing people familiar with the matter.

The move elevates China’s push for semiconductor self-sufficiency as the country’s tech industry confronts restrictive US export controls on advanced chips and chipmaking equipment.

The package under consideration ranges from 200 billion ($28 billion) to 500 billion yuan ($71 billion) in subsidies and financing, and Chinese government officials are still hammering out which companies will benefit and how much each sector will receive, according to Bloomberg.

If rolled out at the top end, China’s proposed spending would dwarf comparable efforts worldwide and surpass funding the U.S. government has committed for the semiconductor sector through the Chips and Science Act. The investment will function separately from Beijing’s existing subsidy program for the chip industry, according to Bloomberg, and signals the government’s determination to reduce its reliance on foreign tech products such as Nvidia’s chips.

4.
OpenAI Releases New Flagship AI Model
By Stephanie Palazzolo Source: The Information

OpenAI on Thursday released a new flagship AI model, GPT-5.2 as the company tries to improve its ChatGPT chatbot and enterprise sales against rivals Google and Anthropic.

The company said that the new generation of models are better at workplace-related tasks including creating spreadsheets, building presentations, writing code, using tools and handling complex, multi-step projects. GPT-5.2 Thinking is the company’s first model that performs at or above a human expert level on GDPval, an OpenAI evaluation that measures AI performance on real-world tasks across domains such as law, finance and engineering.

The company also said that GPT‑5.2 Pro and GPT‑5.2 Thinking are especially optimized for scientific and mathematical work. GPT-5.2 is around 40% more expensive than OpenAI’s last flagship model, GPT-5.1, according to publicly available pricing.

The release comes less than two weeks after OpenAI leaders declared a “code red” within the company, telling employees that they had to refocus resources on improving ChatGPT after Google made technological strides in developing its Gemini 3 model. On Thursday, OpenAI CEO Sam Altman said on CNBC that Gemini 3 “has had less of an impact on our metrics than maybe we feared.”

5.
Nvidia To Host Summit On Data Center Power Shortage
By Anissa Gardizy Source: The Information

Nvidia plans to host a private summit next week with startups focused on solving data center power problems that could hold up the development of artificial intelligence, according to several people who were invited.

The event, at Nvidia’s Santa Clara, Calif., headquarters, signals how energy shortages are impacting companies building facilities filled with Nvidia’s power-hungry artificial intelligence server chips. Executives from power and electrical engineering startups are expected to attend the summit, including firms in which Nvidia has taken equity stakes. The firms sell products ranging from software to physical power equipment technologies.

A spokesperson for Nvidia declined to comment.

6.
Exclusive: Wealthfront IPO Could Triple Tiger Investment
By Katie Roof Source: The Information