YAY! My Claude for Business skool is now running! and crushing it! We have already trained 27 people in just 3 weeks! I am soooo pumped. Thank you to everyone for all the help. Our last client was an $80k solution for Physical Therapists for their 25k clients and they LOVE it. Now we are empowering individual business managers to 8X their day by getting rid of al their bottlenecks and it's just nuts. The execs are taking the 2 day course and it's weird to see them "giddy" about the stuff they are doing ... lol one CEO said "my folk are going to think I went to mars and back when I show them this dashboard" - she takes ladies off the street and helps them create a cleaning business and is doing 1.2M <- we just gave her the team she could find because "it's SO hard to find the right people" hahah Seriously, this ocean is SO blue it's not even funny. We sell the 2 days at $2500 and a landscaper told me " I just saved more than the price of this course by lunchtime on day 1 !! Ask me how much I love hearing that. #onltyquittersfail p.s. thank you <-- even if all you did was read and like this post ;-) Maybe it helps.

Just signed an agreement with a Multi-Naitonal ITIL School with over 55 locations and 12 on sales staff to sell our Claude For Office Curriculum exclusively! TLDR: they will be pitching it to all their clients private and gov and I pick up from the registration on. They keep 40% - We will be their only source of AI training. So basically, I just hired a sales team (with a client base) to pitch my CFO and all I have to do is deliver the 2 day course! Sales Team Meeting Set to present the sales package and Webinar To their list 2 weeks from now. The exciting part, beside the potential 6 fig sales of course, is that I am truly looking forward to those exec "getting it" - We have passed the need for anyone to be a techie now. If you can speak (or type) your ideas in a way that another person would understand, combined with your current knowledge of your work and, tadah - the perfect combo for a massive shift in productivity when you blend those 2 ingredients in an Anthropic Blender ;) Neway, just wanted to share that it's not just AI and Internet Math and some of us ARE making a clean and enjoyable living transitioning folks in the new electricity era . #onlyquittersfail

My father bequeathed me a stack of journals he wrote while recovering from TB at the end of World War II, he was documenting his time in the Wehrmacht, and what it was like on the Eastern Front, where millions died. I have spent decades searching for a way to get these translated. They were written in an old German cursive, so it was tricky. I woke up at 3 in the morning and couldn't go back to sleep when I had the realization that maybe AI can help with their translation. I started scanning some of the first pages and shared them with Claude. Within seconds, my mind was blown... I have a free translator now that can help me write the book, my dad wanted to see me produce! I want to strategize a way now to be able to dedicate hours and hours to the scanning and find payment. I'm thinking AI might be able to suggest a way or even do marketing for donors/patrons to this historical project.

📝 TL;DR Anthropic just published a real world audit of AI agents and the results are surprisingly grounded. Most agent actions still have a human in the loop, most actions are reversible, but the frontier of higher risk and higher autonomy is already showing up. 🧠 Overview Anthropic analyzed millions of agent tool use events across two sources, Claude Code sessions and tool calls made through its public API. Instead of guessing how autonomous agents are, they measured what people actually let agents do, how often humans step in, and how risky those actions look in practice. The big takeaway is that autonomy is not just a model feature, it is co created by the model, the user, and the product design. 📜 The Announcement Anthropic introduced a practical way to score “risk” and “autonomy” for agent actions on a 1 to 10 scale, then clustered similar tool calls to see where real usage sits today. They also examined Claude Code behavior in long interactive sessions to understand when agents stop and ask for help, and when humans interrupt. Headline numbers from public API tool use suggest most activity is low risk and supervised, but there is still meaningful experimentation at the edge, especially in security, finance, and medical data workflows. ⚙️ How It Works • Risk and autonomy scoring - Each tool call is scored from 1 to 10 for risk and autonomy, then grouped into clusters to see which types of actions sit at the extremes. • Most actions are supervised - About 73% of tool calls appear to have a human in the loop in some way, and about 80% appear to have at least one safeguard like restricted permissions or approval requirements. • Irreversible actions are rare - Only about 0.8% of actions look irreversible, such as sending an email to a customer, most actions can be undone or corrected. • Claude Code self limits autonomy - In Claude Code, the agent increasingly pauses to consult the user as tasks get harder, suggesting the model is trained to ask clarifying questions when uncertain.

Trying out a new prompt - would love your feedback - hopefully it's obvious enough that it's for SBM and it's a course on Claude. lol Thanks for the help! #onlyquittersfail