Marla AI for Pro Se Plaintiff’s

Coding a Federal Case: The Multi-AI War Room

The case- Born Vs. AbbVie, Inc – 1:25-cv-12737 – has been weighed in on -AbbVie’s motion to dismiss has been granted WITHOUT prejudice, and we have one more chance to amend. Although it’s bleak, at least we can finally be honest about our methods and be transparent. I am my own pro se plaintiff and I’ve been using AI—and it’s gotten me this far, so already it shows that people can save thousands of dollars. I’ll likely still lose, but so would my former lawyer. I won’t say anything disparaging, but I would have lost the case and tens of thousands of dollars. This way I just lose the case and get to publicize it.

I think what we may be seeing -and this is why I’m showing this to lawyers and people on LinkedIn -is that we might be hitting the limits of what LLMs can do today.

So I’m doing an experiment. I am continuing on with my usual process, which is using multiple AIs and- painstakingly, one paragraph at a time- working on a response in the next week. I spend about 4 hours per paragraph. I ask for legal textbook excerpts, and I really try to learn the “legal logic layer” and I never put in something I don’t understand. In federal litigation, a Rule 12(b)(6) motion is basically a compiler checking your source code for fatal errors. The judge pointed out the exact syntax errors in my legal theory- specifically, how trying to anchor a state-law claim to federal investor-protection language created a terminal conflict that crashed the case. Amending isn’t about rewriting the story; it’s about cleanly divorcing the internal deception I uncovered from the federal preemption trap, keeping the focus entirely on a breach of corporate honesty and the Illinois Whistleblower Act.

The bedrock of my research is my custom Marla-AI.ai chatbot, which I loaded with over 50 pages of raw case history. I uploaded every single court docket entry, alongside a 30-page master narrative I wrote from memory right after I was fired—documenting everything from the architectural defects of ARCH to the 8,000 tickets submitted by frustrated scientists.

With that data securely vectorized, I use ChatGPT’s Canvas tool as my builder to rapidly prototype the raw factual layout under a strict 8-page limit. Then I spend hours on each paragraph – learning the logic, researching every idea – I then feed that output to Gemini and start all over. I tell it to roleplay as a Harvard Professor and rip it to shreds. More hours. More back and forth – it’s what AI teachers refer to as “having the AI’s argue with each other.” So I will say to ChatGPT “Gemini says your wrong on XYZ and we’re going to lose.” Finally, I pass the draft to Claude for a brutal judicial audit to catch any remaining logical gaps before throwing the final version past a couple of incredibly smart people in my inner circle for a human reality check. It’s a workflow that allows a single individual to match the horsepower of a massive corporate law firm. To survive this far is a major accomplishment for whistleblowers everywhere.

10 Prompts You Can Ask Your Marla Chatbot About My Case

To let your inner circle explore the massive wealth of data you’ve uploaded, they can use these specific questions to prompt your chatbot:

1. The Core Defect: “What were the actual technical flaws with the ARCH platform that caused Kathryn to raise red flags in the first place?”
2. The Scientists’ Frustration: “What did the 8,000 internal help tickets reveal about how AbbVie’s chemists and biologists actually experienced the software?”
3. The “Metrics” Scheme: “How exactly did the ‘Excellence Awards’ program work, and how was it used to fake user adoption metrics?”
4. The “Holly” Tool: “What was ‘HOLLY,’ why did Kathryn build it, and how did AbbVie management react when she demonstrated it to leadership?”
5. The Retaliation Timeline: “Can you trace the exact timeline from Kathryn’s first formal report to HR/Ethics in March 2025 to her termination in September?”
6. The Judge’s Critique: “What were the main reasons Judge Jenkins granted the motion to dismiss the previous complaint on July 8, 2026?”
7. The Corporate Mismatch: “What are some specific examples of the mismatch between AbbVie’s public statements about AI and the reality Kathryn observed on the ground?”
8. The Warning Signs: “What performance metrics or praise did Kathryn receive before she was suddenly placed on a Performance Improvement Plan (PIP)?”
9. The Whistleblower Strategy: “Why is shifting the case to the Illinois Whistleblower Act (IWA) a better legal strategy than focusing on Sarbanes-Oxley (SOX)?”
10. The Human Story: “Based on Kathryn’s post-termination narratives, what was the personal and professional turning point that made her decide she had to stand up to corporate compliance?”

Marla BattleBot is here: Chat with HML: Born Vs. AbbVie, Inc – 1:25-cv-12737