Hi everyone — I’m Craig. I work primarily in real estate these days, but my background is a bit of an eclectic adventure. I’ve been a paid stand-up comic, a geophysicist for a major oil and gas company, an insurance brokerage owner, a real estate brokerage owner, and I even had one of my designs featured on the cover of Architectural Digest. I’m based in beautiful Montana, where I’m trying to spend more time playing the piano and I'm not spending nearly enough time fishing. I’ve been exploring AI for a while now. I use ChatGPT regularly, and I even have an AI platform that handles my more tedious social media tasks (because apparently you have to show up online to stay “relevant”). At this stage of life, I’m closer to 70 than 60, and I’m far more interested in spending my time doing meaningful things rather than chasing relevance. I’m here to learn how AI can genuinely support my business and free me up to focus on the things I enjoy. Looking forward to learning from everyone and contributing where I can.

📝 TL;DR OpenAI just told the world that its upcoming AI models could be powerful enough to help launch serious cyberattacks, including finding zero day exploits and breaking into well defended systems. At the same time, it is building new tools and guardrails so those same models can supercharge cyber defense, not just offense. 🧠 Overview A new internal risk assessment from OpenAI classifies its future frontier models as likely to pose a high cybersecurity risk. The company says these systems might eventually help attackers discover new software vulnerabilities or automate complex intrusions, something that has traditionally required elite skills. Rather than hiding this, OpenAI is publicly flagging the risk and announcing new security programs to keep defenders ahead. 📜 The Announcement This week, OpenAI published an update warning that the cyber capabilities of its next generation models are advancing fast enough that it is planning as if each new model could reach high levels of offensive capability. In practical terms, that could mean helping someone develop working zero day exploits or orchestrate sophisticated enterprise or industrial hacks. To counter that, OpenAI is investing heavily in making the same models useful for defenders, creating new access programs, and setting up an expert council focused on frontier risks. ⚙️ How It Works • High risk classification - OpenAI now expects some upcoming models to reach a point where they could materially help with serious cyberattacks, not just generate phishing emails or basic scripts. • Offensive scenarios - The company specifically calls out the possibility that models could help develop new remote exploits, or assist with complex intrusions against real world targets like companies or infrastructure. • Defensive tooling - In parallel, OpenAI is training models to be stronger at security tasks such as auditing code, finding vulnerabilities, and suggesting patches so defenders can move faster too.

If there’s one mindset that separates those who succeed with AI from those who stay stuck, it’s this: the willingness to take imperfect action. So many people hesitate to use AI because they think they need to “understand it” first. They wait for the perfect tutorial, the perfect tool, or the perfect prompt. But AI doesn’t reward perfection. It rewards curiosity. Perfection slows you down. Action teaches you faster than any guide ever could. ------------- The Trap of Waiting to Be Ready ------------- When you first start using AI, it’s easy to overthink everything. What should you ask? How specific should you be? What if the result isn’t good? That hesitation comes from fear: fear of wasting time, fear of doing it wrong, fear of not being “techy enough.” But AI isn’t a test. It’s a conversation. You learn by engaging with it, not by observing from the sidelines. Every time you take action, you teach yourself something new about how AI thinks and how to make it work for you. Progress begins when you stop waiting to be ready and start exploring. ------------- Why Mistakes Are the Shortcut ------------- Here’s the secret that nobody talks about: every AI expert you admire got there by messing up. They wrote bad prompts, misunderstood results, and wasted time on the wrong tools. But they kept going. AI rewards iteration. The faster you make mistakes, the faster you find patterns that work. 1. Start messy - Don’t wait for the perfect prompt. Ask simple questions and build from there. 2. Learn by adjusting - If a result isn’t right, clarify, reframe, and try again. 3. Capture what works - Save your successful prompts and build your own library over time. Each misstep becomes a stepping stone. What feels like failure today turns into intuition tomorrow. ------------- The Power of Small Wins ------------- One of the best ways to build momentum with AI is to create small, achievable wins. You don’t need to master automation or create perfect systems overnight. You just need to prove to yourself that AI can make life a little easier right now.

AI or no AI — just the one that made you say, “Damn, that actually works.” 😄 Curious to hear what everyone’s been cooking up 👇