The Morning Download: AI Governance Gets Teeth

What's up: China’s AI ecosystem accelerates; predictions on how the world of work will change; scenes from this month's AI-Palooza

Good morning. Not too long ago, much of AI governance lived in white papers, voluntary best practices and spongy phrases like "responsible AI." As WSJ Leadership Institute's Isabelle Bousquette reports, some entities are starting to write statutes with teeth.

New York Gov. Kathy Hochul on Friday signed the Responsible AI Safety and Education (RAISE) Act, positioning the state as one of the most assertive AI regulators in the U.S.

The details. Starting Jan. 1, 2027, any company with more than $500 million in revenue that develops a large AI system will have to publish and follow protocols aimed at preventing critical harm from the AI models and report any serious breaches or else face fines. The law also establishes a new office within the New York State Department of Financial Services for enforcement.

Interesting timing. The law was signed shortly after Donald Trump issued an executive order aimed at blocking state-level AI regulation.

Assembly member Alex Bores, the bill’s sponsor, told the WSJLI that states cannot wait for Washington to act. “We need to actually do it at the federal level," he said. But at the same time, "we can’t just be stopping people from taking action to protect their citizens.”

Guidance, but also complexity. For CIOs, state-led efforts introduce the challenge of navigating differing disclosure timelines, definitions of “critical harm” and enforcement structures. Operationally, the RAISE Act goes even further than California’s SB 53 by compressing the time AI developers have to disclose safety incidents from 15 days to just 72 hours.

China’s AI and semiconductor push is accelerating as capital markets reopen

Chinese AI chip maker Shanghai Biren Technology, one of China’s “Four Little Dragons” in the graphics-processing-unit sector, seeks to raise up to about US$623 million in a Hong Kong initial public offering, WSJ reports.

Hong Kong is experiencing an IPO revival. Generative AI startup MiniMax Group filed its documents with the city’s exchange operator after passing the Hong Kong Stock Exchange’s listing hearing. Another large-language-model developer, Knowledge Atlas Technology, better known as Zhipu AI, is also racing toward an IPO in Hong Kong.

China's Moore Threads Technology, a chipmaker founded and helmed by a former Nvidia executive, introduced a new architecture, Huagang, following a blockbuster IPO, Bloomberg reports. Huashan chips will target Nvidia’s Hopper and Blackwell lines, as investors back domestic champions amid U.S. export curbs.

Baidu is teaming up with ride-hailing platforms Uber Technologies and Lyft to begin trials for its self-driving vehicles in the U.K., WSJ reports.

AI tools are fueling a new wave of teenage founders who can build startups faster and younger than ever.

Investors tell the Journal that AI has lowered technical barriers, accelerated learning, and pushed startup ages downward.

Consider Nick Dobroshinsky, who runs BeyondSPX, an AI-generated equity research platform with 50,000 monthly users, built largely by prompting Anthropic’s Claude alongside OpenAI’s ChatGPT and Google’s Gemini. He is 15-years-old.

The Conference on Neural Information Processing Systems, or NeurIPS, once a small and proudly dorky academic gathering has become something more like AI-palooza, WSJ reports.

Sterling Anderson, a former Tesla executive who later co-founded autonomous-trucking startup Aurora, has quickly risen at General Motors after joining as global product chief this summer, WSJ reports. Now overseeing vehicles, software, and AI, Anderson is viewed internally as a dark-horse successor to CEO Mary Barra.

OpenAI improved profitability on paid products, boosting internal “compute margins” to about 70% by October, up from 52% late last year, the Information reports. 

Dina Powell McCormick has resigned from Meta’s board, effective immediately, according to an SEC filing, CNBC reports. Meta will not replace her and may explore a future advisory role. A former Trump and Bush administration official and ex-Goldman Sachs executive, she joined the board in April. Meta now has 14 directors.

A JetBlue flight and two others faced fuel emergencies after an experimental SpaceX Starship rocket exploded Jan. 16, raining debris over the Caribbean and creating an unexpected no-fly zone, WSJ reports. FAA documents show the risk to aircraft was greater than publicly disclosed, compounded by delayed notification from SpaceX.

Five experts look ahead and imagine a future that looks totally different. Read the full story in the WSJ. Highlights below.