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
Anthropic just split the AI industry in half
Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.
Referenced Links:
Anthropic Announces Claude 5 Model Family
TechCrunch: Anthropic's Two-Tier AI Launch Strategy
SWE-Bench: Evaluating AI Software Engineering
AI Architect's Daily Briefing June 12
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. Anthropic just launched two models that split the difference between power and responsibility in the most deliberate way we've ever seen. I'm Chuck Getchell, and this is AI Inten what happened, why it matters, what you can do with it. Let's go. Yesterday, Anthropic released Claude Fable 5 for everyone to use, and Claude Mythos 5 for almost no one to use, and that split tells you everything about where AI is heading. Here's what happened. Fable 5 is the new public model. Think of it as Claude's answer to ChatGPT, but with a twist. It's designed specifically for what Anthropic calls agentic tasks. That's tech speak for AI that doesn't just answer questions, it actually does things, multi-step things, complex things, things that take hours or days to complete. Meanwhile, Mythos 5 is the restricted model. Same company, same technology foundation, but cranked up to 11. Anthropic is only giving access to select enterprise partners and researchers. Why? Because Mythos 5 can handle what they're calling long horizon agentic tasks. It can plan and execute multi-day coding projects, it can manage entire software repositories, it can basically act like a semi-autonomous engineer, which is either exciting or terrifying depending on whether you're the one hiring the engineer or being replaced by one. Let me break down what makes these models different from everything else you've tried. Most AI assistants work like really smart search engines. You ask, they answer. You ask again, they answer again. Each conversation is basically starting from scratch. Fable 5 and Mythos 5 work more like, well, like people, they can break down big goals into smaller steps. They can remember what they were doing yesterday and pick up where they left off. They can call other tools and APIs. They can write code, test it, fix it, and deploy it without you holding their hand through every step. The benchmarks are pretty stunning. Both models are scoring at the top of something called SWE Bench. That's a test where AI has to solve real GitHub issues in actual open source projects, not toy problems, real software engineering work that real humans couldn't figure out. But here's where it gets interesting. Anthropic didn't just build one super powerful model and release it to everyone. They built two versions with very different guardrails. Fable 5 has what they call stronger safety controls. It's more likely to refuse dangerous requests. It explains its limitations better. It gives companies more tools to configure safe usage in their own apps. It's powerful, but it's been deliberately constrained. Mythos 5 has all that power, plus a lot more, but access is restricted because, in Anthropics' words, its level of autonomy and power demands extra safeguards to avoid misuse. Think about that for a second. We now have AI companies saying out loud, some of our technology is too powerful for regular people to use. This isn't some distant sci-fi scenario. This is happening right now, and it's happening because we've crossed a line from AI that helps you work to AI that can actually do the work. So, what does this mean for your life? Let's start with your job. If you work in any kind of knowledge work, writing analysis, project management, basic coding, Fable 5 can now handle multi-step versions of those tasks, not just draft an email for you, but research the background, write the email, format it properly, and keep track of follow-ups, not just suggest code fixes, but read through entire repositories, understand the architecture, write the fix, test it, and document the changes. For people in software engineering, this is a pretty direct challenge. These models can now do a lot of what junior and mid-level developers do every day. Some companies will use this to make their existing engineers more productive, others will use it as justification to freeze hiring or reduce their technical teams. But here's the flip side. If you're not technical, Fable 5 makes technical skills more accessible than ever. Marketing people can now automate their reporting with custom scripts. HR folks can build internal tools without learning to code. Operations managers can create workflows that tie together all their different systems. That's empowering, Vasujari. But it also means your boss might expect you to do a lot more with the same amount of time. The bigger picture is about autonomy. We're moving from AI assistants to AI agents, things that can work on their own for hours or days at a time. The kind of AI that can manage your email while you're sleeping, or write and deploy code while you're in meetings or analyze months of data and present conclusions without you checking every step. That's incredibly powerful. It's also incredibly risky if it goes wrong. Here's what you can actually do with this today. First, if you have access to Claude, try Fable 5 as a project co-pilot, not just a question and answer bot. Give it multi-step tasks. Ask it to plan a six-week content strategy, then draft the first week's posts, then create a tracking spreadsheet. Tell it to break down complex projects into manageable pieces and actually draft the deliverables. The key is thinking in workflows, not individual questions. Instead of write me a blog post, try here's my business, here's my audience, here's my goal, plan a content strategy, identify the best topics, draft the first three posts, and create a publishing calendar. Second, if you run a business, start thinking about which of your processes could be completely automated, not just made faster. Actually, hand it off to an AI agent. Customer support workflows, lead qualification, basic bookkeeping, inventory management, social media scheduling and response. Fable 5 can design these automations for you and even write the scripts or configure the no-code tools to make them work. Third, if you work for a larger company, pay attention to internal AI policies, especially if your employer gets access to something like Mythos 5. Ask questions about what data the AI can access, what tasks it's allowed to perform, and how its work gets reviewed. These agents are powerful enough to touch sensitive information and make real business decisions. You want to know how that's being managed. Which brings me to my bigger point about all of this. The most important thing isn't the technology itself, it's the access model. Anthropic just showed us the future of AI deployment. The really powerful stuff won't be available to everyone. There will be tiers, public models with guardrails, restricted models for enterprises and researchers, and probably even more restricted models for governments and defense contractors. That's probably smart from a safety perspective, but it also means the benefits won't be distributed equally. Companies that can afford access to mythos level agents will have massive advantages over companies stuck with the public versions. Countries with access to the most advanced models will outcompete countries without that access. This is why you need to stay ahead of this curve personally. The gap between AI-enabled people and everyone else is about to get much wider. The gap between AI-enabled businesses and traditional businesses is about to become a chasm. The good news is that even the limited public version is incredibly powerful. Fable 5 can do things that would have seemed impossible just two years ago. You don't need access to the restricted models to transform how you work and what you can accomplish. But you do need to start using these tools seriously, not just for fun experiments or quick tasks. For real work, real projects, real value creation. The companies building AI agents aren't waiting for permission or perfect conditions. They're moving fast and building the future, while most people are still trying to understand the present. Don't get left behind because you were too careful to start. The age of AI assistance is ending. The age of AI agents has begun. And unlike assistants that wait for your instructions, agents are going to reshape everything whether you're paying attention or not. 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.
