Muckrake.ai / Muckipedia is a personal project developed by Steve Hargadon that employs sophisticated Large Language Model (LLM) prompts to analyze historical narratives and current events for potential deception, cognitive vulnerabilities, and overlooked questions. The project emerged from Hargadon's growing awareness that "history, as we tell it, is really the history of power and control" and his desire to understand "what really happens in world events versus how we portray them--which is often not accurately."

Origins and Development

The conceptual foundation for Muckrake.ai developed through several formative experiences in Hargadon's background. His time living in Brazil during high school exposed him to dramatically different perceptions of world events, particularly regarding the United States, which differed significantly from what he had been taught. This international perspective was reinforced by his later reading of "Confessions of an Economic Hitman," which further expanded his understanding of how events are portrayed versus their underlying realities.

During college, Hargadon witnessed the Beech-Nut apple juice scandal, where the company had been selling fake apple juice as real for several years. This event sparked his curiosity about how "so many people who would have been in the know didn't protest (there was one whistleblower)," introducing him to questions about institutional knowledge and silence.

A pivotal influence came from visiting Caen's Centre for History & Peace near Normandy, France, where Hargadon encountered a section devoted to propaganda. There he learned about Edward Bernays and his application of his uncle Freud's theories of the subconscious to manipulate political opinions and purchasing decisions. Hargadon notes that after reading Bernays' book "Propaganda," "one cannot see the world in the same way."

Methodological Framework

The immediate catalyst for developing the Muckrake.ai framework was Hargadon's investigation into the claimed 2015 epidemic of Zika virus-related microcephaly in a small area of Brazil. Using the AI system Grok, he conducted what he describes as a "deep research project" that he believes "uncovered the true cause of the problem--an untested larviside being used without rigorous implementation standards."

From this investigation, Hargadon developed what he characterizes as "a sophisticated prompt that I could give to a Large Language Model (LLM)" designed to identify three key elements:

• Signals of deception
• Reasonable questions that could be asked about an event
• Known human reasoning and cognitive vulnerabilities often exploited in propaganda and advertising

Technical Implementation

The project manifests in two primary forms. The foundational framework and prompt are available at www.muckrake.ai, where Hargadon has posted the prompt and descriptive materials. Building on this foundation, he worked with the LLM Manus.im to create www.muckipedia.com, an interactive site that applies the framework to analyze specific events as prompted by users.

Muckipedia operates by making API calls to either Google's Gemini or Grok, and includes functionality allowing users to compare results between the two systems. Hargadon describes the creation of this interactive platform as producing a "WOW" response, indicating the effectiveness of translating his analytical framework into a user-accessible tool.

Philosophical Approach

Hargadon positions his project within broader concerns about "the pathologizing of critical thinking" and what he sees as the prevalence of "actual conspiracies" in historical events. He explicitly states that his motivation includes addressing "how history is portrayed, how it is often the result of actual conspiracies."

The project also reflects Hargadon's belief in "finding your problem" and using AI as a tool for meaningful work, as he discusses in "Output Shaping: A New Way to Think About the Ethics and Use of AI for Content Creation." For Hargadon, his identified problem centers on "understanding history, how it is portrayed, how it is often the result of actual conspiracies, and the pathologizing of critical thinking."

Limitations and Purpose

Hargadon acknowledges fundamental limitations in his approach, stating that he "remains convinced that LLMs aren't built for 'truth.'" Rather than claiming to provide definitive answers, Muckipedia represents "an attempt just to open the door to understanding where to look more closely." This positions the tool as a starting point for investigation rather than a final arbiter of truth.

The project was developed partly in response to Elon Musk's Grokipedia project, with Hargadon admitting "a little bit of my motivation was to see if I could do a better job of getting to 'truth'" than Musk's initiative, while acknowledging that effectiveness "depends on what you are looking for."