“Every website AI designs looks the same.” “AI writing all reads the same – ‘that X is the point’ etc.” “AI music sounds so bland.” Yes, of course. A probability engine designed to predict the highest probability is going to converge on the middle of the bell curve. Absent strict guidance about what to do…
Summary In today's episode, I walk through how to build a proactive AI agent that identifies wasted ad spend and redistributes budget without manual prompting. Here's what this means for you. You'll see how documented standard operating procedures become the foundation that lets any agentic system function reliably. You'll also learn these concepts: the 5P…
Claude Haiku is the single biggest weakness in the Anthropic ecosystem now. Haiku is intended as a low cost, fast, and cheap model to do basic work – reading, summarization, web search, etc. And at the time of its release almost a year ago, it did those things reasonably well. But times change. Models change.…
_Almost Timely News: 🗞️ Advice I Would Give to a New PR Practitioner (2026-09-06) :: View in Browser_ The Big Plug ✍️ Enroll in my new course, AI for Writers and learn how to make AI write better. Content Authenticity Statement 99% of this week’s newsletter was made by me, the human. You will see…
“Do we really trust Chinese models that much?”” Lisa asked that question in the Trust Insights Analytics for Marketers Slack group and the short answer is, “we really trust models under our control that much”, which means open weights models, models we download and run on our own hardware or in trusted inference providers. Here’s…
Summary In today's episode, I explain the difference between generative answer optimization and agentic SEO, and walk through what companies must do to prepare their websites for AI agents acting on behalf of users. Here's what this means for you. You can future-proof your web presence by adopting Web MCP, which simultaneously improves accessibility for…
Summary In today's episode, I examine the writing styles of major AI systems and reverse engineer which one writes most like a specific human author. Here's what this means for you. You gain a framework for fingerprinting your own writing with Python metrics so you can pick the AI model that sounds closest to you.…
Summary In today's episode, I walk through how to safely feed unstructured customer feedback like reviews and support tickets into language models to extract marketing insights without compromising data privacy. Here's what this means for you. You'll get a clear three-step approach that protects sensitive customer data while still letting you mine feedback for actionable…
AI is a pay cut for you, the human. It’s tempting to take the easy route and talk about a stagnant job market, or tech companies using AI as a cover for large scale job cuts, but that’s not what I’m talking about here. A decade ago, Katie Robbert and I worked at a PR…
Western #AI companies: “Protect us from Chinese AI companies!” Also Western AI companies: “NO you can’t use our models for cybersecurity” Hugging Face ended up using Z.ai GLM-5.2 to defend itself from OpenAI rogue models: https://huggingface.co/blog/security-incident-july-2026 Whenever you hear a Western politician talking about banning/regulating open weights models, it’s because a broligarch has his hands…
