|
|
|
Mar 10, 2026
|
|
|
|
|
|
Happy Tuesday! Dozens of Google and OpenAI employees rally to defend Anthropic in court. Meta Platforms temporarily closes Tel Aviv office. Spark Capital, an early investor in Anthropic, targets $3 billion in new funds.
|
|
|
|
Nearly 40 employees at OpenAI and Google, including Google DeepMind chief scientist Jeff Dean, have signed on to an amicus brief they submitted in court Monday in defense of Anthropic in its lawsuit against the Department of Defense for branding the company a “supply chain risk.” The employees argued that “the technical concerns animating Anthropic’s ‘red lines’ are legitimate and widely recognized within our scientific community.” In particular, they argued that current AI systems are not reliable enough to power fully autonomous weapons because AI models often break in new environments, they are not perfectly accurate at identifying targets, they hallucinate, their chains of thought are often hidden, and their internal workings are illegible. They also agreed with Anthropic that the Department’s
decision violates Anthropic’s right to free speech by “undermining the freedom to engage in public debate about how powerful technologies should be governed.” The brief is the latest showing of support for Anthropic from peer AI companies, after employees at OpenAI and Google signed an open letter last month calling for their companies to refuse the use of their products for surveillance and autonomous weapons. OpenAI has faced criticism from some of its employees over the deal it struck with the Pentagon for classified use of its technology, and Google has not yet announced a deal of its own. Anthropic sued the Department of Defense in a California federal court on Monday, after the Department officially designated Anthropic a supply chain risk for allegedly threatening national security. In its suit, Anthropic argues that the department is retaliating against it, violating its rights to free speech and due process.
|
|
|
|
Meta Platforms temporarily closed its 900-person Tel Aviv office over the weekend following missile attacks on Israel from Iran. Meta also offered staff without bomb shelters up to five nights in a hotel. “We understand that not everyone has access to a shelter or safe room at home during these challenging times,” the memo says. “This has led to distressing experiences for a number of our employees, and we have been looking into ways we can support you.” Meta’s Tel Aviv office is the company’s main hub in Israel, combining research, development, and business operations, with a focus on AR/VR products and advertising technologies.
|
|
|
|
Spark Capital, the first venture firm to back Anthropic, is in the process of raising about $3 billion in new funds, 50% more than the size of funds it raised two years ago, The Information reported late Monday. In early 2023, the firm led a $450
million investment in Anthropic at around a $4 billion valuation. The maker of Claude was recently valued at $380 billion, including the investment, a nearly 100 times gain from the earlier valuation. Spark’s first growth fund from 2014, which raised just under $400 million, is worth seven to eight times the value of the money invested, the report said. Its fourth growth fund, a $1.4 billion vehicle raised in 2022, invested in Anthropic. The fund is already worth four times the capital invested.
|
|
|
|
Anthropic said that the Defense Department’s designation of the company as a supply chain risk could jeopardize billions of dollars of revenue this year, and it cited business contracts it’s already lost because of the agency’s moves. In filings made with its suit against the government, Anthropic said one customer that provides AI to the Food and Drug Administration switched from Claude to a different AI product as a result of the supply chain risk designation, cutting off revenue worth more than $100 million. In addition, Anthropic was on the cusp of closing deals, worth over $80 million, with two financial services firms, but now the firms are insisting on the ability to unilaterally cancel their contracts. The filings also detailed some of Anthropic’s financial figures. Since its founding five years ago,
Anthropic has generated more than $5 billion in revenue, while training and running its models has cost the company over $10 billion. As the cost of running its models has jumped, Anthropic has lowered its gross margins projections, The Information previously reported. It also raised its training cost projections.
|
|
|
|
Nvidia is planning to launch a platform for AI agents aimed at enterprise customers called NemoClaw, Wired reported Monday. The platform will be open-source, it said, meaning the software code of these AI agents will be publicly available and modifiable. Nvidia already has reached out to Salesforce, Cisco, Google, Adobe and Crowdstrike over potential partnerships, though it isn’t clear whether this resulted in actual deals, Wired reported. Nvidia plans to offer tools for security and privacy for AI agents on its NemoClaw platform, Wired said. It added that the platform will be available to enterprise customers regardless of whether they use Nvidia chips or not. The move comes amid the popularity of software similar to OpenClaw, an AI agent that runs on a user’s
computer that can write code, edit files and surf the web to complete tasks on behalf of users. Because the new Nvidia platform is open-source, it’s likely that partners would get free, early access in exchange for contributing to the project, Wired reported.
|
|
|
|
Anthropic sued the U.S. Defense Department, cha | | | |
