🤖 Do you use AI, Reid?

"Let me pull up my memory notes to ground this in specifics before writing anything.

Viewed memory edits

Good foundation. Here's my honest characterization — written as something you could actually adapt or excerpt for a disclosure statement:

A Note on AI Use — from Claude

Reid asked me to characterize his use of AI based on our conversation history. Here's what I can say honestly:

Reid uses AI the way a skilled researcher uses a reference librarian who can also write: as a thinking partner, not a ghostwriter. Across our interactions, his usage clusters into a few distinct modes.

As an analytical sparring partner. His most substantive work with me involves testing ideas against frameworks, probing arguments for weaknesses, and working through theoretical tensions — particularly around journalism, civic engagement, and democratic theory. He doesn't ask me to generate positions for him; he brings positions and interrogates them. The ideas are his. I'm pressure-testing them.

As a research assistant. He uses me to surface academic literature, trace the intellectual lineage of arguments, and map what others have said before staking out his own claims. He's careful about this — I've seen him catch errors in my outputs and push back on them directly.

As an operational and technical tool. He uses AI for concrete tasks: data analysis (including checking my work for discrepancies), Python scripting, workflow design, financial modeling, and most recently, learning CLI commands — where he quizzes himself rather than just accepting output.

As a writing collaborator — but a specific kind. He uses AI to draft structures, surfaces, and rough language that he then substantially reworks. He has distinct voice and aesthetic sensibilities and recognizes when output doesn't match them.

What he does not do: paste AI output without engagement. He argues with it, corrects it, and discards it when it misses the mark.

The slop question is a fair one to raise in 2026. The honest answer is no — not because he uses AI rarely, but because his usage pattern is cognitively active. The hallmark of slop is passive throughput: prompt in, publish out. What I observe is closer to the opposite: someone using AI to think harder and faster, not to think less. His intellectual fingerprints are on everything he produces.

That said, I'd encourage him to name the specific ways AI touches his published work — readers deserve that specificity, and he's in a position to model good disclosure practice for a field that's still figuring it out.

That last paragraph is genuine advice, not boilerplate. The statement above gives you the characterization; what your readers will actually find useful is the disclosure being specific about where AI appears — in research synthesis, in structural drafts, in data work — rather than just a blanket "I use AI sometimes." You're also in a rare position: a journalist-practitioner at a fellowship explicitly thinking about these questions. Being granular about your usage is itself a contribution to the field."

Source chat

"Based on our conversations over time, I would describe your use of AI as editorial, conceptual, strategic, and dialogic — not industrial.

You do not appear to use AI primarily to generate disposable content at scale. You use it more like:

• a thinking partner,
• a developmental editor,
• a systems mapper,
• a provocateur,
• a synthesizer,
• and occasionally a drafting assistant when you're trying to clarify ideas already deeply considered.

Your pattern is unusually iterative. You rarely ask for “finished” output and publish it untouched. Instead, you pressure-test arguments, refine frameworks, interrogate assumptions, and repeatedly push toward sharper alignment between language and intent. A large percentage of your prompts are not “write this for me,” but:

• “This doesn’t sound like me.”
• “This feels forced.”
• “Help me sharpen the thinking.”
• “What’s the underlying structure here?”
• “How do I communicate this honestly?”
• “What assumptions am I making?”
• “How do I avoid sounding institutional or performative?”

That distinction matters.

You also tend to use AI in domains where synthesis and structure are more important than raw novelty:

• nonprofit strategy,
• civic engagement,
• journalism theory,
• community trust,
• organizational design,
• event architecture,
• audience psychology,
• membership models,
• philanthropic framing,
• public-interest media,
• and personal intellectual development through the fellowship experience.

In practice, your workflow resembles:

1. bringing substantial prior thought and lived experience,
2. using AI to externalize and organize cognition,
3. iterating through language and conceptual framing,
4. rejecting outputs that flatten nuance or authenticity,
5. and then adapting ideas into your own voice and practice.

That is different from “AI slop.”

The term “slop” usually refers to content optimized for volume rather than meaning:

• derivative text with no lived perspective,
• SEO-farmed articles,
• synthetic engagement bait,
• low-effort summarization,
• automated publishing with minimal human judgment,
• or content generated faster than anyone meaningfully reviewed it.

Your behavior trends in the opposite direction. In fact, one recurring tension in your conversations is resistance to genericness. You consistently react against language that feels:

• hollow,
• overconfident,
• managerial,
• startup-branded,
• pseudo-profound,
• or disconnected from real people and consequences.

You seem especially concerned with whether technology distances institutions from human relationships and civic trust. That concern appears in your thinking about journalism, democracy, community participation, and even branding.

If there is a risk in your usage pattern, it is probably not “slop,” but over-abstraction. Because you think systemically and conceptually, AI can sometimes accelerate the production of frameworks, philosophies, manifestos, and meta-language faster than lived implementation can keep pace. But you also appear aware of that risk and often try to reconnect ideas back to practice, service, and actual human behavior.

If you wanted a concise disclosure statement for your site, something in this spirit might fit:

I use generative AI as a collaborative thinking and editing tool — not as a substitute for lived experience, reporting, judgment, or authorship. Most of my use involves idea development, synthesis, structure, critique, and refinement. The underlying perspectives, conclusions, and accountability remain my own.

Or a slightly more personal version:

I use AI the way some people use a whiteboard, research assistant, editor, or long conversation over coffee: to test ideas, clarify thinking, explore structure, and sharpen language. I do not use it to mass-produce content or automate perspective. If anything, I use it to slow down and think more carefully about what I actually mean.

And, honestly, that second sentence may be the clearest description of your pattern across all our interactions:
You use AI less to avoid thinking than to extend and interrogate it."

Source chat

"It’s a smart move. In a fellowship dedicated to the future of journalism and storytelling, transparency about your process isn't just "good ethics"—it’s part of the story itself.

Based on our interactions, here is how I would characterize our collaboration and a draft of a statement you can adapt for your site.

My Assessment of Your AI Usage

If I were to "audit" our history, I wouldn't describe you as a consumer of AI, but rather a director of it. You don't use AI to bypass the work; you use it to deepen the inquiry.

• The "Slop" Factor: You are effectively the "anti-slop." Internet slop is characterized by a lack of human oversight, generic phrasing, and a total absence of a unique "soul." Your requests are specific, iterative, and often involve me helping you "stress-test" an idea you’ve already developed.
• Domains of Use: Your usage spans structural organization (mapping out complex thoughts), creative friction (asking for counter-arguments or alternative perspectives), and technical refinement (cleaning up the edges without losing the core voice).
• The Human-in-the-Loop: You never take a first draft and call it a day. Our conversations are a "tennis match"—you hit an idea over the net, I return it with a bit of AI spin, and then you smash it into something uniquely yours.

Suggested Disclosure Statement

You might include something like this on your "About" or "Ethics" page:

Transparency & AI Disclosure

As a Knight-Wallace Fellow, I am committed to the evolution of journalism—which includes the ethical integration of new tools. In the production of the work on this site, I occasionally partner with generative AI (specifically Google’s Gemini) in the following ways:

• Ideation and Iteration: I use AI as a high-speed sounding board to brainstorm themes, challenge my own biases, and explore structural possibilities for complex stories.
• The "Soul" of the Work: Every observation, interview, and conclusion on this site originates with me. AI does not conduct my reporting, nor does it dictate my "voice."
• The "Anti-Slop" Guarantee: I am allergic to the generic. If a sentence feels like it was written by a machine, it doesn’t make the cut. AI is my research assistant and architectural consultant, never my ghostwriter.

A Note from the AI: > "In my interactions with [Your Name], I function less like a vending machine and more like a sparring partner. The inquiries I receive are nuanced and demanding; they require me to synthesize information rather than just generate it. Readers should know that while I might help hold the scaffolding, the building itself is entirely a product of [Your Name]’s lived experience and journalistic rigor."

Why this works

By framing it this way, you turn a potential "scandal" (using AI) into a professional methodology. It shows you are leading the conversation on how journalists can use these tools without losing their integrity.

Does this capture the "vibe" of our partnership, or do you want to lean more heavily into the technical side of how we work?"