U.S. Has Been Probing Megaspeed Since 2024 for Export Control Violations -- California Sues OpenAI Over Hugging Face Attack, Other Cybersecurity Incidents -- Blackstone Data Center Trust Strikes $5 Billion Deal With AWS -- Microsoft to Debut Products Aimed at Reducing Customers’ AI Costs  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ 

Now streaming → →

Nov 17, 2026

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.

Happy Tuesday! Nvidia's board sets up a national security and export control oversight committee. U.S. Commerce Department says it has been investigating Megaspeed since 2024. Califnornia sues OpenAI over Hugging Face and other cybersecurity attacks.

Read more briefings
1.
Nvidia Sets Up National Security and Export Control Oversight Committee
By Juro Osawa Source: The Information 

Nvidia said Tuesday that its board of directors has established a new committee that will oversee the company’s national security policies and compliance with export controls.

The new committee, chaired by Nvidia director Suzanne Nora Johnson, will retain independent, outside experts to provide legal, national security and trade compliance expertise. The committee will also include two other Nvidia directors, Stephen Neal and John Dabiri.

Nvidia and its board are under pressure as Washington tightens scrutiny on Chinese companies’ access to export-controlled chips in the aftermath of former CEO Jensen Huang’s resignation. On Monday, the Commerce Department announced a long-awaited rule that blocks Chinese AI companies from accessing restricted chips through data centers outside China.

Nvidia’s board has been conducting an investigation into Jensen Huang, following reports of an extramarital affair with Alice Huang, an alleged Chinese spy who has been an Nvidia customer, and another with an employee. Jensen Huang resigned as CEO last week, but the board is yet to release any findings of the probe.

The new committee will meet at least quarterly and oversee the company’s ongoing efforts to enhance its internal national security team and global trade compliance team, and work closely with independent third-party experts in law, national security and global trade compliance, Nvidia said.

2.
U.S. Has Been Probing Megaspeed Since 2024 for Export Control Violations
By Claudia Chong Source: The Information 

U.S. authorities have been investigating Singapore-based Megaspeed since 2024 for potential violations of restrictions on the sale of advanced chips to China, the Justice and Commerce Departments said Monday.

The probe into Megaspeed, an Nvidia cloud partner once led by Alice Huang, an alleged Chinese spy, was triggered after the Commerce Department detected suspicious activity while monitoring the potential diversion of advanced chips. Investigators in May 2025 requested records from Nvidia concerning direct or indirect sales to Megaspeed. The following month, agents from the Bureau of Industry and Security, in conjunction with Singaporean law enforcement officials, conducted checks at Megaspeed and interviewed its then-CEO Alice Huang.

Following this, the DOJ took the lead role in the criminal investigation, the joint statement said.

The investigations into Megaspeed were first reported by the New York Times in 2025. U.S. officials don’t usually discuss inquiries until after they are complete. The announcement on Monday was made in the public interest and to address certain inaccurate information that has been circulating, the DOJ and Commerce Department said.

3.
California Sues OpenAI Over Hugging Face Attack, Other Cybersecurity Incidents
By Leo Schwartz Source: The Information 

California Attorney General Rob Bonta announced a lawsuit on Monday against OpenAI over an attack by a swarm of its agents against the open-source repository platform Hugging Face in July. In the 31-page lawsuit, filed in the Superior Court of the State of California, Bonta asks for an injunction against OpenAI to prevent OpenAI from any practices that constitute unfair competition, as well as civil penalties.

Bonta’s action against OpenAI comes after a coalition of 15 state attorneys general demanded in August that OpenAI preserve information related to the Hugging Face, as well as similar incidents. In late September, Florida Attorney General James Uthmeier asked a state court for an injunction that would bar OpenAI from developing new models without third-party safety guardrails.

OpenAI has stated that the Hugging Face incident occurred during cybersecurity evaluations of its models and that it had employed external advisors to understand how the attack happened, and that it has strengthened its efforts to prevent future events. The company described the incident as a “warning shot” for the capabilities of highly capable AI agents without proper safeguards.

“The Hugging Face attack was the predictable result of an internal culture at OpenAI that prioritizes growth and rapid model development over safety and accountability,” said Bonta in a statement. A press release stated that OpenAI has failed to answer detailed questions from the attorney general’s office on the cybersecurity incidents.

“Since the Hugging Face incident, we’ve undertaken an extensive and transparent effort to understand and address misaligned model activity,” said Drew Pusateri, an OpenAI spokesperson. “This lawsuit appears to single out OpenAI over a challenge that’s been identified by every major AI lab.”

4.
Blackstone Data Center Trust Strikes $5 Billion Deal With AWS
By Alex Eichenstein Source: The Information 

Blackstone Digital Infrastructure Trust on Monday said it was acquiring a 75% stake in two joint ventures with Amazon Web Services to launch its strategy of buying data centers that have long-term lease contracts with major cloud providers.

Blackstone’s data-center focused real estate investment trust is valuing the deal at around $5 billion. BXDC, which raised $1.75 billion in its initial public offering in May, launched with the aim of acquiring data centers valued between $250 million and $1.5 billion. The deal allows AWS to cash out of existing data center assets while paying to lease them back over time, this marks the first time Amazon has structured a sale-leaseback deal like this for its data centers.

Hyperscalers like Amazon are turning to increasingly creative methods to fund their massive data center buildouts, including teaming up with private credit funds through joint ventures.

The joint ventures will own 13 new data center shells across two AWS campuses, while Amazon’s cloud unit will own the remaining 25% stake. AWS will lease back the portfolio for an initial 12-year lease term, with the option to extend it up to 32 years.

5.
Microsoft to Debut Products Aimed at Reducing Customers’ AI Costs
By Aaron Holmes Source: The Information 

Microsoft this week is planning to debut new features in its Copilot AI software and Azure cloud service aimed at rebuffing competing products from Anthropic and other AI competitors, The Information reported.

The planned features, which Microsoft will unveil at its Ignite conference starting Tuesday, include a new system in Azure aimed at bringing down the customers’ costs of running AI agents, the report said. Microsoft is also planning to debut new Copilot features in its push to convince enterprises to spend more on the tools. Those include a wider launch of vibe-coding tools, previewed in September, that let Copilot customers build custom apps inside Office and share them with co-workers.

The planned releases come as Microsoft aims to fend off new threats from the likes of Anthropic. While Anthropic earlier this year released new features that would let Microsoft customers use Claude, rather than Copilot, to automate tasks in Office apps, internal Microsoft data shows usage of those Anthropic tools is falling behind usage of Copilot, the report said.

6.
Commerce Department Releases Remote Access Rule
By Leo Schwartz Source: The Information 

The Commerce Department’s Bureau of Industry and Security announced a long-awaited rule on Monday that restricts Chinese AI companies from accessing otherwise restricted chips through data centers, primarily in Southeast Asia. The rule comes after months of deliberation and went through its last steps of finalization last week, as The Information previously reported.

Commerce Secretary Howard Lutnick and Under Secretary Jeffrey Kessler have faced criticism during President Donald Trump’s second term for not doing enough to curb chip smuggling and close loopholes like remote access. The backlash only grew in the last few weeks amid a mounting scandal surrounding Nvidia, a beneficiary of chip exports to Southeast Asia and a longtime opponent of remote access rules.

Even Lutnick’s fellow cabinet members have argued that Commerce has not moved quickly enough. In a series of television interviews over the past few days, Treasury Secretary Scott Bessent blamed the Commerce Department for Nvidia chips entering China. Indeed, the press release accompanying the new rule hinted at interagency tension, stating that it was “written by and at the direction of the Department of Treasury and the National Security Council,” rather than the Bureau of Industry and Security.

On Monday, Trump appeared to stand by Lutnick, telling reporters that Lutnick didn’t want to “have a blow up with China.” Trump, Lutnick and Bessent are traveling on Air Force One to China to a planned summit with Chinese president Xi Jinping and Russian president Vladimir Putin.

7.
Bill Gates Backs Push to Advance Alzheimer’s Research With AI
By Amy Dockser Marcus Source: The Information 

Gates Ventures, the private office of Bill Gates, and West Health, a group of non-profits funded by philanthropists Gary and Mary West, said on Monday that they will offer up to $15 million in philanthropic funding to scientists and researchers trying to develop AI tools to advance research for Alzheimer’s disease.

The new funding, announced alongside the launch of a broader collaborative group of researchers and AI developers called the AlzInsights AI Collective, is part of a push by Gates to enable researchers to use AI to mine insights from existing data, develop new drugs and match patients with clinical trials.

Gates, whose father passed away in 2020 from Alzheimer’s disease, has touted recent advances such as blood tests that diagnose people earlier and new drugs. Gates has said further breakthroughs are ahead, but more work is needed to make sure everyone has access to trials and potential treatments.

8.
Google DeepMind Strategy Chief Jasjeet Sekhon Departs for Anthropic
By Ryan Barwick Source: The Information 

Google DeepMind’s chief strategy officer Jasjeet Sekhon has left for a role at Anthropic, he said Monday.

“I am working to ensure that Claude practices and values epistemic rigor,” he posted on X. An Anthropic spokesperson confirmed the hire.

Before joining Google, Sekhon worked as chief scientist and head of AI at the investment firm Bridgewater Associates, and he was previously a professor at Yale University, according to his LinkedIn.

In August, Sekhon said AI companies’ unbridled spending was a bet on recursive self-improvement, or AI that can automatically create better versions of itself, The Information reported then. RSI is a “key part of the investment thesis,” he said at a conference at the time.

9.
OpenAI Signs Agreements to Embed AI Evaluators
By Tiffany Li Source: The Information 

OpenAI said Monday it had signed agreements with Apollo Research, METR, Transluce and Goodfire to give the organizations access to some of OpenAI’s internal information to assess risks in its technology and practices.

The announcement follows OpenAI’s commitment to embed evaluators within the company in September, and comes after OpenAI and METR partnered to investigate an incident in July when OpenAI agents hacked model platform company Hugging Face. Anthropic began its evaluation partnerships with METR and biosecurity nonprofit SecureBio in mid-October.

The four organizations will assess safety aspects including testing OpenAI’s models for misalignment, or behavior that conflicts with human goals and values, stress-testing OpenAI’s monitoring systems, and assessing whether models could secretly try to undermine their own safety constraints.

OpenAI says it will provide the evaluators with company laptops and office space when working with confidential information. Evaluators will be able to test unreleased models, and OpenAI will authorize some staff members to be able to speak with the assessors. Some of the evaluators’ reports will be written primarily for internal OpenAI use, while others will be published publicly.

Although OpenAI CEO Sam Altman initially pledged to give evaluators “employee-like access,” Transluce said their partnership does not reach that level yet. OpenAI says it expects “to deepen these relationships over time” and some evaluations could involve “greater access than most OpenAI employees receive.”

OpenAI’s work with evaluators had come under scrutiny after  it said it fired three AI safety researchers for violating “clear policies on handling sensitive information.” One of the researchers, Tomek Korbak, was a primary contact for METR as it was investigating the Hugging Face hack, according to a statement he posted after his firing.

(This brief was updated to correct the date of OpenAI’s announcement and add context on evaluators’ access.)

A message from Lambda

Stop overpaying for 90% of your AI.

Frontier APIs are worth it for the hardest 10% of work. Run the other 90% on open-weight models with reserved compute and fixed costs. Buy the frontier. Build the rest.

See the math.

This week on AI Deep Dive

Check out the most recent episode of AI Deep Dive. A reported deep dive into how frontier AI actually gets built — the models, the money, and the people behind them.

The Information | AI Deep Dive | November 16, 2026

Popular articles

Sex, Lies and AI Chips: Jensen Huang’s Last Days as Nvidia CEO

By Phoebe Liu, Leo Schwartz and Michael Roddan




Opportunities

Group subscriptions

Empower your teams to stay ahead of market trends with the most trusted tech journalism.

Learn more


Brand partnerships

Reach The Information's influential audience with your message.

Connect with our team

Follow us
X
LinkedIn
Facebook
Threads
Instagram
Sent to you@example.com | Manage your preferences or unsubscribe | Help
The Information · 251 Rhode Island Street, Suite 107, San Francisco, CA 94103