The first week of April crystallizes one theme: AI has become so powerful and valuable that it now demands unprecedented security, capital, and regulation. Anthropic's new Claude Mythos preview (codename Capybara) is so capable the company restricts it to a handful of partners through its Project Glass Wing security initiative — after the model uncovered thousands of zero-days, including a 27-year-old OpenBSD bug. Meanwhile, OpenAI, Anthropic, and Google form the Frontier Model Forum to fight model theft via “adversarial distillation.” Funding hits historic highs: $300B raised globally in Q1, with $242B flowing to AI and OpenAI's $122B round setting an all-time record. A USC Dornsife study warns that shared chatbots are shrinking humanity's cognitive diversity. And the White House unveils a national AI legislation framework that clashes with state laws — even as it cuts Anthropic from government business, prompting a lawsuit.

Anthropic locks down its most powerful model

Claude Mythos (codename Capybara) is so capable that Anthropic restricts it to 12 partners via Project Glass Wing. In testing it uncovered thousands of zero-days, including a foundational 27-year-old bug in OpenBSD.

Rival labs unite against model theft

OpenAI, Anthropic, and Google formed the Frontier Model Forum to combat “adversarial distillation” — Chinese firms like DeepSeek copying models via API, a practice costing US labs billions of dollars annually.

AI funding smashes records in Q1 2026

Startups raised a record $300B globally in Q1, with $242B going to AI. OpenAI's $122B round is the largest venture deal ever; Anthropic, xAI, and Waymo added tens of billions more.

Billions pivot toward physical AI infrastructure

Investment is surging into data centers, robotics, and semiconductors as AI's appetite for power and hardware grows. Seed funding rose 31% year-over-year, but late-stage megarounds now dominate.

Study warns of shrinking cognitive diversity

A USC Dornsife study finds that as billions rely on the same chatbots, collective creativity shrinks. LLMs favor linear reasoning, quietly standardizing how humanity thinks.

White House unveils national AI legislation

A new federal framework pushes age verification, protects residential electricity rates, and limits developer liability — clashing with states like California. Trump cut Anthropic from government work, and the company is suing.