We’ve spent the last couple of years treating generative AI like a vending machine. Select a task. Insert a prompt. Retrieve a product. And to be fair, in many legal and professional contexts that’s exactly the right frame: accuracy and precision matter and “creative” output in payroll or billing codes is usually just a polished
The End of the Magic Wand: Why 2026 Demands Resilience Prompting
For more than two years, lawyers have been told that success with generative AI depended on writing better prompts and a search for the perfect “magic wand” prompting formula. That was the wrong lesson. The real change in 2026 is not found in the model itself, but in the professional posture required to use it.
Prompting or Negotiating? A Systems Design Lesson for Legal AI
I had a long session recently with a public genAI tool that taught me something more important than the topic I started with.
The lesson was not about whether the model was “smart enough.” It was about control. At a certain point, I realized I was no longer simply prompting an LLM. I was negotiating
Moving Beyond Prompts to Protocol-Governed AI
I have just posted a trio of new research white papers to SSRN. These represent the latest output from the Kennedy Idea Propulsion Laboratory and the culmination of my work over the last month to move AI beyond “utilitarian drift.” This is the cycle of incremental efficiency gains that ultimately generates no transformative insight.
The Inquest: Trading the AI Idol for Human Investigation
In 1980, I wrote a senior thesis paper called “Imagination: A Romantic Ideal.” My investigation then was a critique of the German and English Romantics who, in their zeal to undo the “damage” of Enlightenment Reason, merely erected a new idol: The Imagination.
Through a concrete analysis of Keats and Poe, I discovered a truth
The Holmesian Leap: Solving the “Wicked Problem” of Access to Justice with AI
- The Problem: The self-represented litigant (SRL) crisis is not a simple “information gap” but a “wicked
The Operational Protocol Method: Systematic LLM Specialization Through Collaborative Persona Engineering and Agent Coordination
This paper has been published and and a PDF of it is available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5397903
The Operational Protocol Method: Systematic LLM Specialization Through Collaborative Persona Engineering and Agent Coordination
By Dennis Kennedy
August 19, 2025
Kennedy Idea Propulsion Laboratory Working Paper No. 2025-01
License: This work is licensed under the Creative Commons Attribution-ShareAlike 4.0
Is the Legal Profession Ready to Win the AI Race? America’s AI Action Plan Has Fired the Starting Gun
The Starting Gun for Legal AI Has Fired. Who in Our Profession is on the Starting Line?
The legal profession’s “wait and see” approach to artificial intelligence is now officially obsolete.
This isn’t hyperbole. This is a direct consequence of the White House’s new “Winning the Race: America’s AI Action Plan.” After spending the last
HyperCard: Illustrating My AI Prompting Approach
I’ve been working on organizing and optimizing my AI prompts and prompting methodologies. I noticed that I have two major categories of approaches.
The first I call “complex, structured prompting.”
Google Gemini describes that, and I think accurately, as “a systematic, engineered way to interact with AI. It’s about building a personalized ‘cognitive operating system’
Celebrating the 22nd Birthday of DennisKennedy.Blog
Well, here we are again. Another February 15th and another blawgiversary for DennisKennedy.Blog. It’s hard to believe it’s been twenty-two years. since I wrote that first bold and triumphant announcement, echoing Babylon 5: “And so it begins . . .” I genuinely thought I was late to the blogging party back in 2003. Turns out