🤖 New York Just Pulled the Emergency Brake on AI — and That's Not Even the Biggest Story Today
Welcome to Daily Inference, your daily dose of the most important AI stories shaping our world. I'm your host, and today is July 14th, 2026. We've got a packed episode — from governments putting the brakes on AI infrastructure, to a landmark lawsuit between two of tech's biggest names, to real workers losing real jobs to AI right now. Let's get into it.
But first, a quick word from our sponsor. If you've ever wanted to build a website without spending hours wrestling with code or design tools, check out 60sec.site. It's an AI-powered tool that lets you create a beautiful, functional website in under a minute. Seriously. Head to 60sec.site and see for yourself.
Alright, let's start with what might be the most significant regulatory move in AI infrastructure we've seen from a U.S. state. New York Governor Kathy Hochul has signed the nation's first statewide moratorium on new hyperscale data centers. For up to one year, no new environmental permits will be issued for data centers drawing more than 50 megawatts of power. To put that in perspective, that's enough electricity to power tens of thousands of homes. The stated goal is to give New York time to craft proper regulations that protect residents from spiking energy costs and environmental damage before Big Tech plants another server farm in their backyard. And this might just be the beginning — the state legislature has passed an even stricter bill, covering facilities above 20 megawatts, that's still waiting on Hochul's signature. What's really interesting here is the timing. Communities across the country have been pushing back against data center construction for years, worrying about noise, water use, and strain on local power grids. New York just became the first state to actually legislate a pause. As AI models grow larger and more energy-hungry, expect this tension between the AI boom and local infrastructure to become one of the defining policy battles of the decade.
Speaking of government and AI, there's a troubling story out of Washington. The Department of Housing and Urban Development used AI tools developed by DOGE — the Department of Government Efficiency — to help shape housing policy. That alone raises eyebrows. But here's the kicker: when journalists filed a public records request to find out exactly how that AI was used, HUD withheld the documents, citing a legal privilege that, according to legal experts, simply doesn't exist. This is the intersection of two already fraught conversations — AI in government decision-making, and government transparency. When an algorithm influences who gets housing assistance or how policy gets written, the public has a right to know how it works and whether it's been tested for bias. Hiding behind a made-up legal shield is not a good look, and it's likely just the beginning of these kinds of accountability battles as AI gets embedded deeper into government operations.
Now, let's talk about Apple versus OpenAI, because this lawsuit is wild. Apple has filed a trade secrets case against OpenAI containing some genuinely jaw-dropping allegations. According to the complaint, when Apple employees were being recruited to join OpenAI, the AI company's hardware head allegedly asked job candidates to bring along unreleased product samples and prototype components to their interviews. The lawsuit centers on three individuals, including Tang Tan, a 24-year Apple veteran who served as VP of the Apple Watch before departing for OpenAI in 2024. Apple is accusing OpenAI of systematically extracting confidential design knowledge, spying on hardware prototypes, and tricking a trusted partner into performing a proprietary product design technique. This case is unfolding at a fascinating moment — Apple is simultaneously doubling down on its own AI ambitions with a revamped Siri that's now the backbone of iOS 27, available in public beta right now. The new Siri is described as less of a voice assistant and more of an everything tool woven into the entire iPhone experience. So Apple is fighting OpenAI in court while racing to compete with it in the market. The irony is thick.
Let's shift to a story that hits closer to home for a lot of people. Twelve nurses at Montefiore hospital in the Bronx were laid off and replaced with AI-powered software. These weren't junior staff — one of them, Marilyn Shuler, had worked as a utilization review nurse for 39 years. Her job involved reading patient charts and coordinating with insurance companies over coverage decisions. Now an algorithm does it. The New York State Nurses Association says the hospital violated a contract the nurses had only recently won through a strike. This story lands at the same moment economists are putting a more precise timeline on AI's labor market impact — the question is no longer if AI displaces workers, but when and how fast. And for these 12 nurses, that timeline already ran out. What makes this particularly significant is that healthcare, long assumed to be relatively insulated from automation because of its human complexity, is clearly not immune.
On the technical front, Mistral AI has released something genuinely impressive: a model called Robostral Navigate. It's an 8 billion parameter model that can guide a robot through complex real-world environments using only a single standard RGB camera — no fancy laser sensors, no depth cameras, just the kind of camera you'd find in a decent smartphone. The model achieves a 76.6 percent success rate on a standard navigation benchmark, which is a strong result for a model this size. Why does this matter? Because one of the big barriers to deploying robots in everyday environments has been the cost and complexity of their sensor arrays. If you can get strong navigation performance from cheap, commodity hardware, that dramatically lowers the cost of building capable robots. Combine this with the broader trend of agentic AI systems getting better at operating in the physical world, and you start to see how the robotics industry could accelerate faster than most people expect.
Finally, a quick look at the funding landscape, which remains absolutely frothy. Video generation startup PixVerse just raised 439 million dollars, pushing its valuation north of 2 billion. And Nous Research, the company behind the Hermes family of AI agents, is reportedly in talks to close at least 75 million dollars at a 1.5 billion dollar valuation, led by Robot Ventures with participation from Union Square Ventures. The money keeps flowing, the models keep improving, and the real-world consequences keep multiplying — that's the AI moment we're living in.
That's a wrap on today's Daily Inference. If you want to go deeper on any of these stories, head to dailyinference.com for our daily AI newsletter — we break down the most important developments every single day. And remember to check out today's sponsor, 60sec.site, for AI-powered website creation that actually works as fast as it sounds. Thanks for listening, stay curious, and we'll see you tomorrow.
