|
|
|
Aug 27, 2026
|
|
|
|
|
Supported by
|
|
|
|
|
|
Happy Thursday! Meta agrees to pay up to $18 billion to settle a lawsuit alleging its apps harmed children. Nvidia agrees to buy Hugging Face for $12.9 billion. SoftBank is in talks to buy a majority stake in 1X Technologies in a deal that would value the humanoid robot startup at about $6 billion.
|
|
|
Gibson Dunn, a Los Angeles-headquartered law firm well-known for its litigation practice, has hired tech lawyer Jonathan Ashtor from New York law firm Paul Weiss, according to a release reviewed by The Information. The hire is the latest in a series of senior moves among the country’s largest law firms. Ashtor is joining as a partner and co-chair of Gibson Dunn’s technology transactions practice group. Ashtor was most recently a partner and served as Paul Weiss’s co-chair of its global artificial intelligence group. In the past year, he advised on high-profile transactions such as Amazon in its $10.8 billion acquisition of Globalstar and IBM’s $11 billion acquisition of data streaming firm Confluent. This is the latest in Gibson Dunn’s aggressive push to recruit from rivals. Last month, it hired William Savitt, one of the co-chairs of rival Wachtell, Lipton, Rosen & Katz, as well as five other partners in its litigation department. This year, Gibson Dunn advised SpaceX on its $1.25 trillion merger with xAI, its $60 billion acquisition of coding agent startup Cursor and SpaceX’s $86 billion IPO. Big Law is locked in fierce competition for top lawyers, and hundreds of law firm partners have made lateral moves this year. Prominent law firms are also increasingly shifting their focus on technology transactions beyond M&A, such as IP licensing, joint ventures and various forms of asset-backed and leased-backed financings.
|
|
|
Meta Platforms agreed to pay up to $18 billion to help settle a lawsuit filed by states alleging Meta’s social media apps harmed children, bringing a costly conclusion to one of the biggest legal threats facing the social media giant. Under the proposed settlement, which requires a judge’s approval, Meta agreed to a series of changes to its Instagram platform and other services, including default daily limits and nighttime blocks for teenage users, enhanced age assurance measures to prevent children from accessing Meta’s platforms, and the creation of new tools to help parents protect their children online. Meta also agreed to pay each of the 52 states and territories involved varying amounts in two parts. One part is to be paid in annual installments over a decade totaling about $12.7 billion, or about 70% of the total, Meta said in a statement. The remaining 30%, totaling about $5.3 billion would be paid out on the condition that YouTube and TikTok make similar changes to their services that Meta has agreed to, and that those companies each pay an amount matching the 30% figure. (Attorneys general in their statements provided a slightly smaller total of up to about $17 billion.) The settlement agreement covers a bipartisan group of 52 attorneys general across U.S. states, territories, and Washington, D.C. If approved, it would conclude a trial that began last week at a federal courthouse in Oakland, Calif., involving a subset of 29 states. The payments amount to a sizable sum even for a company that reported $60 billion in profit last year, but the number is far smaller than some projections for what Meta could have faced: Meta attorneys at one point said the states were seeking $1.4 trillion in penalties, and the states put the number at closer to $200 billion, according to Reuters. Meta still faces other lawsuits, including from individuals. It has already lost cases including one in California brought by a young person and one in New Mexico that resulted in a total of $942 million in penalties.
|
|
|
Nvidia has agreed to buy Hugging Face, the GitHub-like repository of open-source AI models, for $12.9 billion, roughly 80 times the startup’s forward revenue, after deal talks kicked off when another suitor came calling, The Information reported. Salesforce, an investor in the company, has been among the interested parties, the report said. Nvidia is purchasing the startup to shepherd the open-model ecosystem as a counterweight to closed-source model developers such as Anthropic and OpenAI, both of which are building their own AI server chips to lessen their reliance on Nvidia’s. Owning Hugging Face also revives Nvidia’s cloud business.
|
|
|
SoftBank is in talks to buy a majority stake in 1X Technologies, an OpenAI-backed humanoid robot developer, The Information reported late Wednesday. The investment would support SoftBank’s robotics ambition and give 1X more runway to put its soft-bodied bots in customers’ homes. The 12-year-old startup tried last fall to raise $1 billion at a $10 billion valuation, but it raised less than half that target, the report said. Last year, OpenAI and 1X discussed a possible acquisition but the talks didn’t progress.
|
|
|
Nvidia’s revenue rose 106% to $96.2 billion in the three months that ended in July, or 21 percentage points higher than the growth it reported in the previous quarter. The company said growth would cool a bit, to 89.5%, in the current fiscal quarter. But given its recent results, investors are likely to expect the company to blow past that estimate the way it has done in recent quarters, given strong demand signals from cloud providers that represent most of Nvidia’s data center hardware revenues. However, two new issues bubbled up: supply chain constraints will slow growth to 70% in the 12 months ending Jan. 31 2028, CFO Colette Kress said, though that was a significantly higher projection than many investors expected. Second, customers may not be paying Nvidia as quickly as they used to. Its free cash flow fell more than 50% to $21 billion, compared to the April quarter, as the company disclosed an around 50% increase in accounts receivable “due to extended payment terms on large, multi-quarter agreements with certain investment-grade customers,” a term that typically refers to major cloud providers and large customers such as Meta Platforms and SpaceX. Shares rose 5% in after-hours trading after Nvidia announced the growth projection as well as new AI server purchases by Amazon that implied additional revenue of at least $100 billion for Nvidia. When normal trading hours ended Wednesday, Nvidia’s stock was up 11% this year, but it was cheaper than ever: its price-to-earnings ratio was 20.8, down from 26.7 three months ago and 33 a year ago. Nvidia for the first time segmented its data center revenue between “hyperscale” customers and everyone else. Approximately 55% of Nvidia’s data center revenue came from hyperscale customers. This disclosure replaces Nvidia’s previous segmentation, which broke out revenue from networking products like switches and cables versus compute, which mostly consists of AI server chip systems. (Networking has been a fast-growing part of Nvidia’s revenue, but customers frequently buy Nvidia’s networking products alongside its server chips.) Nvidia is also continuing to use its balance sheet to support its customers in different ways. Nvidia disclosed for the first time that it signed large, long-term data center lease agreements worth $20 billion, which it plans to reassign to third parties—presumably its customers. It also disclosed that the value of its equity investment holdings was $99 billion, and that it had committed to making $25 billion in future equity investments. Nvidia has been putting money into AI application firms such as Perplexity, which use its chips, cloud providers that buy its chips, and data center development firms that work with cloud providers. Nvidia previously committed to make a $30 billion equity investment in OpenAI, and it also disclosed $108 billion in credit support on behalf of OpenAI, which plans to lease a large data center facility in Ohio, filled with Nvidia chips, likely starting in 2028. Kress said that while “some will call circular financing, we see it differently” because Nvidia is aiding companies that will “become the largest technology companies in history.” She said that a quarter of Nvidia’s revenues next year would be related to AI labs that Nvidia is supporting financially.
|
|
|
Salesforce stock rose more than 13% in after-hours trading Wednesday after the enterprise software giant reported 11% growth in July quarter earnings and boosted its full year revenue guidance to as high as $46.4 billion from $46.2 billion, meaning revenue could increase as much as 12% in Salesforce’s fiscal year ending next January from the previous year. Salesforce’s July quarter results were generally in line with the company’s projections. The revenue growth was two percentage points slower than the April quarter. The company’s current remaining performance obligation, or the revenue from current customer contracts the company expects to see over the next year, climbed 14% to $33.5 billion, as expected. Notably, Salesforce’s annual recurring revenue from its suite of tools for building and running AI, dubbed Agentforce, continued to move higher, up 240% from the previous year to more than $1.5 billion in the July quarter. That figure—typically defined as the current monthly subscription multiplied by 12—was $1.2 billion in the April quarter. Including Salesforce’s data management service Data 360, which is used alongside its AI tools, AI-related ARR rose around 210% from the previous year to $3.9 billion. Salesforce also said it expanded its partnership with Anthropic, launching a new tool that makes it easy for customers to use their Salesforce accounts through Claude.
|
|
|
Instinct, a startup whose AI assistant connects to users’ applications and performs tasks on their behalf, is raising a Series B at a $2.5 billion valuation in a round led by Index Ventures and Benchmark, said a person with direct knowledge. The round will close in the next couple of weeks with around $250 million raised. Instinct was founded in April by Noah Shin, a former researcher at the AI startup Sierra, and has raised roughly $350 million to date including from Kleiner Perkins, Conviction and Greenoaks, according to a source familiar with the matter. The Wall Street Journal earlier reported on the investment. The startup, which launched its product only a few months ago, has drawn significant buzz in Silicon Valley. “It’s the first AI assistant that can do complex tasks on your behalf and accomplish actions rather than just answering questions, ” said Mike Vernal, partner at Conviction. Investors are also betting on Shin, who worked on machine learning at Northeastern and MIT. Venture capital has poured money into AI assistants that take over your computer and help you with daily tasks after open source personal agent OpenClaw went viral in late January. Increased use of agents has also lifted revenues for model makers like Anthropic and | | |
