AI News Roundup: Stanford Index Reveals China Closing Gap, Anthropic Tops OpenAI in Revenue, Meta Launches Muse Spark

Stanford AI Index 2026: China Closes the Gap to 2.7%

Stanford’s annual AI Index report, released this month, paints a striking picture of the global AI race. The performance gap between the best American and Chinese AI models has collapsed to just 2.7%, down from as much as 31.6 percentage points in May 2023 — despite the U.S. outspending China on private AI investment by a factor of 23 ($285.9 billion vs. $12.4 billion).

China now leads in AI patents (69.7% of global filings), research publications (23.2% of global output), and industrial robot installations at nine times the U.S. rate. Meanwhile, AI capability continues to accelerate: several frontier models now meet or exceed human performance on PhD-level science questions, and on the SWE-bench Verified coding benchmark, performance rose from 60% to near 100% in a single year.

Perhaps most concerning for U.S. competitiveness, the number of AI scholars moving to the United States has dropped 89% since 2017, with 80% of that decline occurring in the last year alone.

Anthropic Surpasses OpenAI With $30 Billion Revenue Run Rate

Anthropic announced this month that its annualized revenue run rate has crossed $30 billion — up from roughly $9 billion at the end of 2025. For the first time, Anthropic has overtaken OpenAI, which currently sits at $25 billion. The company now counts over 1,000 business customers each spending more than $1 million annually, a figure that doubled in less than two months.

To fuel the growth, Anthropic expanded its partnership with Google and Broadcom for approximately 3.5 gigawatts of next-generation TPU capacity starting in 2027, adding to about 1 gigawatt of Google compute already committed for 2026. Separately, Amazon announced it will invest $5 billion immediately in Anthropic, with up to $20 billion in additional funding tied to performance milestones. Investor offers now value Anthropic at $800 billion, more than double its $350 billion valuation from February.

Meta Debuts Muse Spark in “Ground-Up Overhaul” of Its AI

Meta unveiled Muse Spark on April 8, its first flagship model built under Chief AI Officer Alexandr Wang’s newly formed Superintelligence Labs division. The natively multimodal model — code-named Avocado and built over nine months — supports voice, text, and image inputs with tool use, visual chain of thought, and multi-agent orchestration.

Meta also introduced “Contemplating mode,” which orchestrates multiple agents reasoning in parallel, designed to compete with frontier reasoning modes like Gemini Deep Think and GPT Pro. Unlike Meta’s previous open-weight Llama models, Muse Spark is proprietary and currently powers Meta AI across its apps, with WhatsApp, Instagram, and Ray-Ban Meta AI glasses rollouts coming in the weeks ahead.

Meta simultaneously announced AI capital expenditures of $115–$135 billion for 2026, nearly double last year’s spending.

Snap Cuts 16% of Workforce as AI Reshapes Tech Employment

Snap announced it would lay off roughly 1,000 employees — 16% of its workforce — with CEO Evan Spiegel directly citing “rapid advancements in artificial intelligence” as the driving force. It marks the first time Spiegel has explicitly linked AI to staffing decisions, noting that AI now generates more than 65% of Snap’s new code.

The restructuring is projected to strip over $500 million from Snap’s annualized cost base by H2 2026. Markets responded favorably, with Snap shares jumping 7–11% on the news. The cuts add to a broader wave — Oracle plans to cut up to 30,000 employees, and Atlassian, Block, Pinterest, and Salesforce have all trimmed headcounts, with AI efficiency consistently cited as the rationale.

PwC Study: 20% of Companies Capture 74% of AI’s Economic Gains

A sweeping new PwC study of 1,217 senior executives across 25 sectors finds that 74% of AI’s financial gains are going to just 20% of organizations. The distinguishing factor isn’t how much AI a company deploys, but how — leaders use AI as a growth and reinvention engine rather than purely for cost reduction.

Companies with the best AI-driven outcomes are nearly twice as likely to deploy autonomous, self-optimizing AI systems and are increasing the number of decisions made without human intervention at 2.8 times the rate of peers. The single strongest predictor of AI-driven performance is pursuing new revenue from industry convergence, not efficiency gains alone.

NVIDIA and Adobe Bring Agentic AI to Enterprise Marketing

NVIDIA and Adobe deepened their strategic partnership with a new suite of agentic AI tools for enterprise content creation and customer experience. The collaboration brings together Adobe’s new CX Enterprise Coworker platform, NVIDIA’s Nemotron open models and Agent Toolkit, and WPP’s global marketing expertise to enable autonomous agents that generate, adapt, and activate content at scale.

A key highlight is Adobe’s now generally available 3D digital twin solution built on NVIDIA Omniverse and OpenUSD, which lets AI agents automate high-fidelity content creation across markets, formats, and product configurations while maintaining brand compliance and governance.