US AI policy Brief — 2026-06-01 - AI Consultant | Enterprise Agentic AI

Covering developments published in the 72h to 2026-06-01 10:59:20 (+0800).

Top Stories

NBC News · 2026-05-29
Summary: The Illinois House voted 110-0 to send SB 315, the Artificial Intelligence Safety Measures Act, to Governor J.B. Pritzker, who has vowed to sign it. The bill — already passed by the Senate on May 21 — requires large frontier AI developers (those with more than $500M in annual revenue and models meeting a frontier-scale compute threshold) to commission annual independent third-party audits of their safety practices, submit pre-deployment transparency reports, report critical safety incidents to regulators, and implement whistleblower protections for employees. Notably, OpenAI and Anthropic publicly backed the bill, while a broader industry trade group opposed it. The law takes effect January 1, 2028.
Why It Matters: SB 315 is the first US statute to impose mandatory external accountability audits on frontier AI labs — a direct legislative response to the governance vacuum left by the White House’s scrapped federal AI security executive order. It sets a benchmark other states and ultimately Congress may follow, and practically obligates the top handful of AI companies (OpenAI, Anthropic, Google, Meta) to institutionalize safety review processes.
URL: https://www.nbcnews.com/tech/tech-news/illinois-legislature-passes-historic-ai-bill-rcna347191

Fortune · 2026-05-29
Summary: Anthropic closed a $65B funding round valuing the company at $965 billion — surpassing OpenAI’s $852B valuation — and announced that its restricted Mythos-class models will reach all customers “in the coming weeks.” Mythos, whose cyber-offense capabilities (automated zero-day discovery, exploit chaining) previously prompted the White House to block broader access on national security grounds and to draft — then abruptly abandon on May 21 — a frontier model oversight executive order, will now enter wide commercial release alongside Claude Opus 4.8, a newer model designed with enhanced refusal and anti-deception properties.
Why It Matters: The Mythos wide-release announcement directly reopens the federal regulatory question that the scrapped May 21 EO was meant to address: who vets powerful dual-use AI models before they reach the open market, and on what terms? With the White House having backed down from mandatory pre-deployment testing and no federal statute in place, Anthropic’s move puts maximum pressure on Congress and regulators — and validates the urgency behind the Illinois approach of mandating third-party audit frameworks.
URL: https://fortune.com/2026/05/29/anthropic-raises-65-billion-at-record-965-billion-valuation-promises-mythos-ai-model-in-wide-release-in-coming-weeks-releases-claude-opus-4-8/