AI in 10
The most important AI story—explained in 10 minutes.
Every day, I break down the biggest AI story in just 10 minutes - what it is, why it matters, and how you can actually use it. No tech jargon, just AI made simple.
AI in 10
Illinois just made AI safety the law
Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.
Referenced Links:
Illinois Artificial Intelligence Safety Measures Act — Illinois General Assembly
Illinois becomes first state to pass major AI safety law — Axios Chicago
TechNet Statement on Illinois AI Auditing Requirements — TechNet
YouGov: 14% of DIY Investors Now Using AI Instead of Human Advisors — YouGov
Want to go deeper with AI? A community of professionals is learning AI together right now at aihammock.com — show notes, links, tools, and real conversations about how to actually use AI in your life.
Welcome to AI Inten. I'm Chuck Getchell, and every day I break down the biggest AI story in just 10 minutes. What it is, why it matters, and how you can actually use it. Illinois just became the first state in America to make AI safety the law. Not a suggestion, not a guideline, the law. I'm Chuck Getchell. This is AI Inten. What happened? Why it matters, what you can do with it. Let's go. Here's the setup. For the past few years, the biggest AI companies in the world have been publishing what they call safety frameworks. Nice little documents, thoughtfully worded, often very long, and almost entirely voluntary. Nobody checked whether they actually followed them. Nobody had to. It was basically the honor system for trillion-dollar technology companies shaping the future. What could go wrong? That era just ended in Illinois. Earlier this week, Governor J.B. Pritzker signed the Artificial Intelligence Safety Measures Act into law. He was joined at the signing by Illinois Attorney General Kwame Raul and the state lawmakers who pushed this bill through back in May. And what they created is something genuinely new in the United States: a legally enforceable set of obligations for large AI companies. Not a pilot program, not a task force, law. So what does the law actually require? Let's break it down into plain English because the legaleese can get thick fast. First, big AI companies, the ones building and deploying powerful advanced AI models, must now publish a risk and transparency framework. Not just a vague we care about safety blog post. A detailed document that explains how they measure what their models can actually do, how they identify dangerous capabilities, and how they respond when something goes wrong. They have to show their work. Which for some of these companies is going to be a significant adjustment from just posting a friendly press release and calling it a day. Second, and this is the one that has the tech industry nervous, the law requires independent annual audits. Third-party organizations not connected to the company have to verify that the AI developer is actually following the safety processes they described every year. This is important because a safety framework you write yourself and never have to prove you followed is not really a safety framework. It's more of a creative writing exercise. Third, companies must have documented procedures for handling serious safety incidents. If one of their AI models causes a significant harm, bad medical guidance, a security breach, something that affects a lot of people, there has to be a process for identifying it, reporting on it, and updating their frameworks in response, no more hoping nobody notices and quietly patching things in the background. The law has a specific focus on what it calls catastrophic risk. That phrase is intentional. Lawmakers weren't just thinking about a chatbot giving bad restaurant recommendations, they were thinking about scenarios where powerful AI could be used for large-scale cyber attacks, for spreading mass disinformation, or for helping bad actors access dangerous knowledge. The kind of stuff that, if it happens, doesn't just inconvenience you. It affects communities, systems, entire sectors of society. Now the tech industry has opinions about all of this. Shocking, I know. A major industry coalition called TechNet has already pushed back on the mandatory third-party audit requirement. Their concern is that outside audits could slow down product releases, create legal liability, and lead to a confusing patchwork of different rules across different states. And honestly, that last concern is legitimate. If every state passes its own AI safety law with different requirements, companies are going to be spending a lot of time on compliance paperwork instead of building things, did you say? But supporters, while including the Attorney General's office, make a fair counter-argument. When you're talking about AI systems that touch hiring decisions, financial recommendations, healthcare, and public safety, voluntary commitments from the companies that profit from those systems may not be enough. At some point, somebody outside the building needs to take a look. That's not a radical idea. We already do that with banks. We do it with pharmaceutical companies, we do it with airlines. The question was always when AI would reach that level of consequence. Apparently, Illinois decided the answer is now. So let's connect this to your actual life. Because if you live in Illinois, this is directly relevant, and if you live anywhere else, it may be coming your way soon. Legal experts are already saying this law could become a template for other states. Think about where powerful AI is already touching your daily life. If your employer uses AI to screen resumes, evaluate your performance, or manage your schedule, there's an AI model behind that. If your bank uses AI to approve or flag your transactions, model behind that. If you've used an app that gives you health information or investment guidance powered by a large AI system, model behind that too. Until now, in most cases, you had no real way to know how those models were built, what risks the company considered, or what happens when they make a mistake. This law doesn't give you a magic window into the code, but it does require that the companies behind those systems document and prove their safety practices. And that creates a paper trail, which creates accountability, which creates pressure to actually get it right. Here's something worth noting if you're in finance or healthcare specifically, as I always say, I'm not sure I'm not a lawyer, doctor, or financial advisor, and you should always talk to a professional for your specific situation. But the direction this law is pointing is toward more transparency in exactly those high-stakes areas where AI errors can do the most damage. The Ugov research released earlier this month actually found that 14% of DIY investors in the US are already using AI tools to guide their investment decisions. 14%. And that number is only going up. Laws like this one are starting to catch up to that reality. Alright, let's talk about the one actionable thing you can do with this information right now. Whether you live in Illinois or not, you now have a new question in your toolkit. And the question is simple: Does this company have a published AI safety framework? You can ask that question of your employer's HR software vendor, you can ask it of your bank, you can ask it of any company that uses AI in a way that affects your life, your money, your job, or your health. Here's how to actually do it. Go to the company's website and search for terms like AI safety or responsible AI, model card or transparency report. Major AI companies like OpenAI, Google, Anthropic, and Microsoft already publish versions of these. Smaller vendors sometimes don't. If you can't find anything, that tells you something. And if you find a document that's all buzzwords and no specifics, that also tells you something. If you work in HR, operations, legal, or procurement at any company that's evaluating AI vendors right now, this is worth adding to your vendor checklist. Ask vendors directly. Do you have a published risk and transparency framework? Do you undergo any external safety review? What is your process when your model causes an error that affects a customer or employee? These are fair, reasonable questions. And the quality of the answer will tell you a lot about how seriously a company takes the technology it's selling you. For those of you who are newer to all of this and not quite sure what a foundation model even is or why you should care about AI audits, my AI Explained course is a 30 video series built specifically for people who are just getting started, no technical background required. It walks you through exactly what these tools are, how they work, and why decisions like the one Illinois just made are going to matter to your everyday life. Worth checking out at AI Hammock. Here is the big takeaway from everything we covered today. Illinois just drew a line in the sand. It said, if you build powerful AI and you want to operate here, you have to be able to prove your safety practices to someone who doesn't work for you. That's a meaningful shift, not a revolution, not a solution to every AI problem, but a shift from trust us to show us. And in an industry that has been mostly running on trust us for the past several years, that matters more than it might seem. That's today's AI Inten. If you want to go deeper and learn AI with a community of people just like you, join us at aihammock.com. I'll see you tomorrow, my friends.
