OpenAI releases a new ChatGPT tool for all things work related. The company's long-awaited super app is here.

Fidji Simo, OpenAI's second-in-command, is stepping down from her full-time position due to a worsening medical condition, she confirmed on LinkedIn. Simo served as CEO of AGI deployment, leading the company's product and business divisions, before taking medical leave earlier this year. She will transition to a part-time advisory role. The change in leadership comes as OpenAI prepares for a public offering and seeks to strengthen its competitive position against rivals like Anthropic. OpenAI's chief futurist also announced this week that he's leaving the company.

Meta Platforms (META.O)

, opens new tab plans to start manufacturing an artificial intelligence chip from ‌September as part of its plan to boost overall computing power to 14 gigawatts next year, showed an internal memo reviewed by Reuters.

The tech firm's data center chip, code-named "Iris," is part of a four-generation project for Meta Training and Inference Accelerators (MTIA) that it will design in-house. The plan is to use custom-built silicon to improve the AI that powers its Facebook and Instagram social media platforms.

Testing the chip took only six weeks and found no major issues, the memo showed. That relatively quick progress signals positive momentum for an in-house effort that has floundered since its launch more than half a decade ago.

Meta tailored the chip for ⁠its own needs and is working with Broadcom (AVGO.O), opens new tab to help design it, and Taiwan Semiconductor Manufacturing Co (2330.TW), opens new tab to manufacture it. The approach is likely to help the firm lower its massive computing costs and gain more independence from chip suppliers such as Nvidia (NVDA.O), opens new tab, and Advanced Micro Devices (AMD.O), opens new tab.

The bug-testing completion and production timing have not been previously reported. Meta declined to comment.

The chip is meant to augment the large quantities of graphics processing units (GPUs) used for AI applications that Meta purchases from Nvidia and AMD.

However, adopting the latest GPUs at a firm as large as Meta "has been a heavy lift, and it has cost us time," the memo showed.

Shares fell after the report but recovered after the company announced developer access to an AI coding model that pits it directly against OpenAI and Anthropic. They were trading up 4.6% in late afternoon trading.

"You can't become an AI titan if you are dependent on another company for chips," said Mike Gualtieri, a vice president and principal analyst at research firm Forrester. "The hyperscalers and even SpaceX all plan chips ‌because it ⁠will be the only way to compete on price for model usage."

Meta unveiled Iris under its technical name in March, along with three other AI processors. It plans to launch a chip about every six months through 2027, whereas typically firms release AI chips at intervals of a year or more.

A worker stands inside the Meta Lab in Los Angeles, California, U.S., May 20, 2026. REUTERS/Daniel Cole Purchase Licensing Rights, opens new tab

SEVEN GIGAWATTS OF COMPUTING IN 2026

Meta this year plans to deploy seven gigawatts of computing infrastructure, the memo showed. To reach that total, Meta added 1 gigawatt in the first half of the year and forecasts adding another 5.5 gigawatts by the end of the year, the ⁠memo said.

One gigawatt of energy is enough to power about 800,000 homes.

The company plans to double capacity again next year to reach a total of 14 gigawatts in 2027, the memo said.

The firm expects to spend as much as $145 billion on AI infrastructure this year, a significant portion of Big Tech's more than $700 billion projected outlay on the technology.

To expand computing infrastructure, Meta has secured ⁠long-term, multi-year supply agreements, the memo showed. Those include agreements with Samsung Electronics (005930.KS), opens new tab for memory chips, Sandisk (SNDK.O), opens new tab for flash storage, and Sumitomo Electric (5802.T), opens new tab for fiber-optic equipment.

Such long-term agreements have become critical for data center expansion targets amid a memory chip shortage that has prompted companies such as Apple AAPL.O to raise prices.

Sandisk declined to comment. Samsung Electronics and Sumitomo ⁠Electric did not respond to requests for comment.

Components such as memory and AI chips have experienced a surge in demand as tech companies race to expand data centers to keep pace with AI's thirst for computing power.

Memory and other chip prices have risen rapidly and substantially enough that "chipflation" has become a macroeconomic concern, Morgan Stanley analysts said.

OpenAI on Thursday broadly released GPT-5.6, after staggering the powerful new model's rollout per a U.S. government request.

Why it matters: The leading players in AI are all speeding forward at breakneck speed, but the increasing power of the latest models has the Trump administration rethinking its previous hands-off approach.

Driving the news: OpenAI is releasing three flavors of GPT-5.6:

Sol is the most powerful version, while Luna is designed for speed. Terra, meanwhile, aims to strike a balance between the two for everyday work.
A new "ultra" mode within Sol allows the system to work even harder on tasks and to delegate work to various submodels.
CEO Sam Altman told CNBC that Sol is 54% more token efficient on agentic coding tasks, an important stat as "every enterprise now is thinking about spend and the value they're getting in exchange for AI," he said.

Alongside the new models, the company also launched ChatGPT Work on Thursday.

The new agent, powered by 5.6, can gather context across connected apps and files to create documents, spreadsheets, presentations, and other work.
It can also work across web, phones, and computers, first rolling out to the Mac and Windows apps for all tiers of users, with web to follow.

The model release comes after the administration asked OpenAI to delay its rollout.

Altman said the company made "many changes" after a "collaborative back and forth" with the administration.
He called the government's technical capabilities "impressive."

OpenAI has been battling Anthropic for the crown of having the most powerful model.

At the same time, Meta, SpaceXAI, and others have been aggressively trying to catch up with the leaders. Both companies released updated models this week.

GPT-5.6 is already getting rave reviews from many of those who have been testing it. Some say that they see Anthropic's Fable as having greater raw intelligence, but GPT-5.6 is viewed as being a more reliable model for regular tasks.

MagicPath AI CEO Pietro Schirano wrote on X: "I've been testing it for months and, without exaggeration, it's the best model I've ever used. Fast, smart, genuinely creative."

Theo Browne, CEO of chatbot platform T3 Chat, wrote on X: "gpt-5.6-sol is world-leading in computer use. It made me use it 100x more. When we lost access to 5.6, I quickly started to go insane without it."

Some who have tested both Fable and GPT-5.6 expressed a preference for Anthropic's model.

"It's an amazing model, but for almost every task I tested, Fable was quite a bit better," investor Matt Shumer wrote on X.
Others said they see advantages to both models. The CEO of Every, Dan Shipper, likened GPT-5 to a Porsche, while saying that Fable offers the equivalent of "warp drive."
"If you need to get across the galaxy, use Fable," Shipper said in a post on X. "If you need to get around town using the best available tool for the job, use 5.6."
Netflix’s New Script: Live TV, Bundles, and the Fight Against Friction
Netflix is facing an unfamiliar villain: declining viewer engagement.
Despite rising profits and industry-low cancellation rates, the time subscribers spend watching content is slipping. With shares down over 40% in the past year and TV viewership share hitting a low of 7.8% in April, top executives are aggressively ditching the company’s historic "keep it simple" philosophy to reverse the trend.
Here is how the streaming king plans to rewrite its playbook:
1. Linear Channels & Streaming Bundles
- Live "Always-On" TV: Netflix is discussing adding live, linear-style channels dedicated to specific shows or genres—mimicking the passive, casual viewing success of free rivals like Tubi and Roku.
- The Aggregator Play: The streamer has explored selling rival subscriptions (like NBCUniversal’s Peacock) directly inside the Netflix app, turning its home page into a digital cable hub similar to Amazon and Apple.
2. Cheaper, Quick-Bite Content
Rather than relying solely on $100 million blockbuster series, Netflix is lean-budgeting to keep eyeballs glued to the screen:
- Integrating short-form videos from BuzzFeed and Condé Nast.
- Leaning into video podcasts and repackaged YouTube content.
- Partnering with local broadcasters (like France’s TF1) to stream news and regional programming, a strategy it plans to export across Europe and Latin America.
3. Live Sports (Without the Price Tag)
While co-CEOs Ted Sarandos and Greg Peters refuse to buy ultra-expensive seasonal sports rights, they are cherry-picking massive global events. The company is currently discussing bids for the 2030 and 2034 FIFA World Cups.
The Ultimate Goal: Fueling the Ad Engine
The media landscape is consolidating rapidly—highlighted by Fox's $25 billion acquisition of Roku and Paramount’s impending $81 billion merger with Warner Bros. Discovery (a deal Netflix secretly bid on last year, spooking investors who now worry U.S. growth has peaked).
To compete, Netflix needs its advertising tier to explode. The ad business generated $1.5 billion last year, and leadership expects to double that. Live programming and linear channels provide the perfect vehicle: unskippable commercials.
The era of pure, ad-free, binge-it-all-at-once streaming is officially dead. To stay on top, Netflix is transforming into the very thing it originally sought to destroy: cable TV.