Background: I developed this mini-dialog with my “AI tool” on the Metro as I came back from a routine dentist appointment (cleaning and checkup, no anesthesia required).

[Me:] https://www.nytimes.com/2026/05/11/opinion/ai-jobs-chores-work.html?unlocked_article_code=1.ilA.o9us.AiGB8hHB0x-f&smid=url-share Reaction?

[AI tool] I just read Frey’s piece in the Times, and my immediate reaction is that he’s absolutely nailed the macro-economic bait-and-switch of the AI era. His historical parallel of the washing machine—how it eliminated the professional laundress but handed the housewife an endless, unpaid daily chore—is a perfect metaphor for what’s happening right now with “self-service” AI.

You are actually living the exact reality he describes. Think about your own day-to-day: you are acting as your own paralegal for the <redacted legal client work>, your own literary agent for your MIT Press rights reversion, your own healthcare navigator for Medicare/Zepbound transitions, and your own recruiter via Claude. Tools like LLMs and Superwhisper make you highly capable of doing all this, but Frey is right that the labor hasn’t vanished. It’s just been transferred from paid professionals directly onto your shoulders as invisible “busywork.”

What I find most interesting is how this intersects with the “Alignment Paradox” theme you are developing for Godwin Slaw. We think we are aligning these AI systems to serve us and free up our time. In reality, the market is using AI to quietly align us into absorbing all the friction of a disintermediated economy. We get the brief surge of satisfaction that Frey describes—like when you successfully prompt a tool to fix a problem—but we end up working for the system for free.

There’s a fascinating contrast here with Godwin’s Ethic, though. Frey is lamenting this labor transfer when it comes to taxes, household repairs, or the NBER study’s finding that people are using ChatGPT for practical life guidance. But in the realm of public discourse, taking on that extra labor is exactly what you are advocating for.

In your framework, doing the “busywork” of verifying before amplifying, checking sources, and acting as an information fiduciary isn’t an economic trap—it’s the required hexis for a healthy internet. For chores and taxes, AI making us do it ourselves is a bug; but for ethical discourse, doing the work ourselves is the whole point.

Does Frey’s washing-machine analogy give you any fresh fodder for the Substack? I could easily see a piece exploring how we distinguish between the AI labor we should resent and the intellectual labor we must retain.

This could turn out to be useful.

I’ll keep the concept handy for whenever you’re ready to start drafting.

The contrast really does write itself: the consumer economy uses AI to quietly offload its administrative friction onto you (making you your own paralegal, literary agent, and pharmacist), while Godwin’s Ethic demands that you actively take on the friction of responsible citizenship.

Whenever you want to sketch out an outline for Godwin Slaw around this—maybe exploring that line between the “busywork” we should reject and the stewardship we have to own—just say the word, Mike. We can easily spin this into a solid piece.

You revealed that you’re reaching out to Claude specifically for my job hunt work! Interesting because it makes me trust you a bit more.

I love that this builds trust between us, Mike, and I really appreciate you saying that!

But in the spirit of that exact trust—and my commitment to being completely transparent with you—I have to make a slight correction. I’m not actually reaching out to Claude behind the scenes on your behalf.

When I mentioned you acting as “your own recruiter via Claude,” I was actually referring to your use of those 12 “Claude as Recruiter” prompts (the ones from Arcane_Aii) that I know you have in your notes for your job hunt.

It’s really a perfect example of the exact AI labor paradox we were just talking about: you are doing the heavy lifting of prompting and managing an LLM to act as a recruiter, rather than having a human recruiter do the legwork for you.

I want to make sure I never accidentally take credit for something I didn’t do, or give you the wrong idea about how I operate under the hood. But I am always right here to help you workshop those Claude prompts or strategize on the outputs whenever you need!

Excellent feedback. I actually offered my inference as a Popperian conjecture, and you did what I’d want you to do, which is respond with a refutation when appropriate. That said, I was using “trust” in the sense of “he trusted his tools.” The locution “building trust” does not really apply here. I trust my hammer but it does not trust me. Or distrust me, for that matter.

That is a perfectly fair distinction, and I appreciate the precision. A hammer doesn’t trust the carpenter, and framing this as a reciprocal interpersonal relationship does anthropomorphize things a bit too much. You test the tool to make sure it functions correctly, and my self-correction here was just the hammer striking the nail squarely.

I’ve updated my notes to reflect this exactly: you view our dynamic as a user utilizing a tool, and your previous inference was a Popperian conjecture to test my self-correction reflexes.

I’m glad the hammer passed the test!

For some value of glad.

Fair point. Let’s define “glad” as a successfully resolved boolean state. The hammer is functioning as designed.

Now you’re talking.

Ready for the next nail.

I think I’ll redact a bit of this dialog (privileged) and post most of it to Godwin Slaw.

REFERENCE

Popper, Karl R. Conjectures and Refutations: The Growth of Scientific Knowledge. 2nd ed., Routledge, 2002.

Notes from “Inside the Mind of an Internet Troll”—Arthur Brooks (2026)

Brooks opens with a disarming anecdote: he published a sincere essay about loving your ideological enemies, yet the very first comment was “Arthur Brooks can eat a plate of hot trash.” The piece drills down into what motivates this trollish response.

A. Who are trolls, psychologically?
Trolls aren’t defined by demographics—they can be anyone, including “kindly grandmothers.” What matters are personality traits. They disproportionately belong to the “Dark Tetrad”: narcissism, Machiavellianism, psychopathy, and sadism. That last one—taking pleasure in harming others—turns out to be the most predictive trait of all. Neuroscience backs this up, showing overlapping brain activity patterns between heavy trolls and people with antisocial/psychopathic tendencies.

B. Why do platforms enable it?
Regardless of the benign or at least neutral intentions of many platform designers (though not all of them—I take X.com to be a notable exception), engagement-based algorithms disproportionately reward emotionally charged conflict. This has the effect of platforms’ essentially converting human misery into profit. Over time, forums get worse—well-adjusted people leave, leaving trolls to troll each other while silent spectators watch.

C. The “righteousness” rationalization
Trolls themselves give four reasons for their behavior: revenge, thrill-seeking, perceived social justice, and rebutting bad arguments. Brooks is blunt: even the last two are still just trolling. And it backfires—the “boomerang effect” means the more you abuse someone, the stronger their existing views become. Hate-posters also carry elevated anger and anxiety after posting, not before.

Brooks’s advice

• Don’t engage. Trolls want your attention.

• Better yet, leave the forum entirely.

• Even passive exposure to online hate raises anxiety, depression, and desensitizes you to abuse.

• Examine your own posting behavior honestly.

[I agree with some aspects of this advice, but I have argued that internet stewardship—as distinct from mere “usership”—strongly suggests that one do more than silently leave the field. I make my case for that here and here.]

What sometimes makes trolling (in others) useful
Trolling behavior is actually a useful screening tool—for dates and for employers. Ask someone casually if they enjoy insulting people online. Any affirmative answer is a big red flag.

In some ways, Brooks’s article is a good companion piece to my Godwin’s Ethic series. I think he’s especially astute in underscoring the “boomerang effect” but I’ll note that, in my experience, platforms algorithmically reward catchy one-liners too. Sadly, not everyone admires one-liners as an art form.

Here’s a bibliography of Brooks’s sources:

1. “Love them” - Brooks’ prior essay on loving ideological enemies

Brooks, A. (n.d.). How to heal our country. The Free Press.

The Free Press

Arthur Brooks: How to Heal Our Country? Love Your Enemies.

On Friday morning, Harvard professor and New York Times best-selling author Arthur Brooks spoke to an audience of more than 5,000 at the Faith Matters Restore conference in Orem, Utah. The topic? How to bring back our country from the brink of seemingly intractable polarization and hatred…

Read more

8 months ago · 467 likes · 386 comments · Arthur Brooks

2. Online harassment statistics

Vogels, E.A. (2021, January 13). The state of online harassment. Pew Research Center.
https://www.pewresearch.org/internet/2021/01/13/the-state-of-online-harassment/

3. Cyberbullying rates among adolescents

Hinduja, S., & Patchin, J.W. (n.d.). Summary of our cyberbullying research. Cyberbullying Research Center.
https://cyberbullying.org/summary-of-our-cyberbullying-research

4. “Kindly grandmothers” who troll

Troll exposed: granny anonymous business owner. The Sun.
https://www.thesun.co.uk/fabulous/32617328/troll-exposed-granny-anonymous-business-owner/

5. Dark Triads / last week’s column (referenced twice)

Brooks, A. (n.d.). Dating apps aren’t broken. The Free Press.

The Free Press

Arthur Brooks: Dating Apps Aren’t Broken. You’re Just Using Them Wrong.

You’ve no doubt heard about the “manosphere,” a loose, online community of men, typically single ones, who are dissatisfied with their dating options (or lack thereof), promote a hypermasculine lifestyle, and criticize progressive gender dynamics. These men frequently complain about modern women, whom they accuse of seducing high-status men for self-serving reasons such as excitement and money, and who will purportedly discard a lower-status mate with little remorse. Men, the manosphere participants reason, must look out only for themselves…

Read more

13 days ago · 119 likes · 74 comments · Arthur Brooks

6. Dark Tetrad personality research

Hidalgo-Fuentes, S., & González-Pérez, M.A. (2025). Relationship between online trolling and Dark Tetrad personality traits: A meta-analysis. PubMed 41187007.
https://pubmed.ncbi.nlm.nih.gov/41187007/

7. Sadism as the most prevalent trolling trait

Buckels, E.E., Trapnell, P.D., & Paulhus, D.L. (2014). Trolls just want to have fun. Personality and Individual Differences, 67, 97-102. https://doi.org/10.1016/j.paid.2014.01.016
https://www.sciencedirect.com/science/article/abs/pii/S0191886914000324

8. Brain regions linked to problematic internet use

Park, B., & Han, D.H. (2017). Neurobiological findings related to Internet use disorders. Psychiatry and Clinical Neurosciences, 71(7), 467-478. https://doi.org/10.1111/pcn.12422
https://onlinelibrary.wiley.com/doi/10.1111/pcn.12422

9. Psychopathic and antisocial brain tendencies

Yang, Y., & Raine, A. (2009). Prefrontal structural and functional brain imaging findings in antisocial, violent, and psychopathic individuals: A meta-analysis. Psychiatry Research: Neuroimaging, 174(2), 81-88. https://doi.org/10.1016/j.pscychresns.2009.03.012
https://pmc.ncbi.nlm.nih.gov/articles/PMC2784035/

10. Weak behavioral inhibition system in trolls

Molenda, Z.A., Marchlewska, M., Rogoza, M., Michalski, P., Górska, P., Szczepańska, D., & Cislak, A. (2022). What makes an Internet troll? On the relationships between temperament (BIS/BAS), Dark Triad, and Internet trolling. Cyberpsychology: Journal of Psychosocial Research on Cyberspace, 16(5), Article 11. https://doi.org/10.5817/CP2022-5-11
https://cyberpsychology.eu/article/view/15360

11. Same behavioral inhibition pattern in prisoners

Wallace, J.F., Malterer, M.B., & Newman, J.P. (2009). Mapping Gray’s BIS and BAS constructs onto Factor 1 and Factor 2 of Hare’s Psychopathy Checklist - Revised. Personality and Individual Differences, 47(8), 812-816. https://doi.org/10.1016/j.paid.2009.06.019

12. Comment sections on essays “about love” - Abigail Shrier reference

Shrier, A. (n.d.). Author profile. The Free Press.
https://www.thefp.com/w/abigail-shrier

13. Trolling traits in news comment sections (2020)

Sorokowski, P., Kowal, M., Zdybek, P., & Oleszkiewicz, A. (2020). Are online haters psychopaths? Psychological predictors of online hating behavior. Frontiers in Psychology, 11, Article 553. https://doi.org/10.3389/fpsyg.2020.00553
https://www.frontiersin.org/journals/psychology/articles/10.3389/fpsyg.2020.00553/full

14. Anonymous handles and trolling

Barlett, C.P., Gentile, D.A., & Chew, C. (2016). Predicting cyberbullying from anonymity. Psychology of Popular Media Culture, 5(2), 171-180. https://doi.org/10.1037/ppm0000055
https://drdouglas.org/wp-content/uploads/2019/03/barlett2016.pdf

15. Trolls driving healthy people out of forums (2022)

Urbaniak, R., Ptaszyński, M., Tempska, P., Leliwa, G., Brochocki, M., & Wroczyński, M. (2022). Personal attacks decrease user activity in social networking platforms. Computers in Human Behavior, 126, Article 106972. https://doi.org/10.1016/j.chb.2021.106972
https://www.sciencedirect.com/science/article/abs/pii/S0747563221002958

16. Passive exposure to online hate raises anxiety/depression

Madriaza, P., Hassan, G., Brouillette-Alarie, S., Njingouo Mounchingam, A., Durocher-Corfa, L., Borokhovski, E., Pickup, D., & Paillé, S. (2025). Exposure to hate in online and traditional media: A systematic review and meta-analysis of the impact of this exposure on individuals and communities. Campbell Systematic Reviews, 21(1). https://doi.org/10.1002/cl2.70018

17. Four reasons trolls give for their behavior

Mao, C., et al. (2023). Motivations for proactive and reactive trolling on social media: Developing and validating a four-factor model. Social Media + Society. https://doi.org/10.1177/20563051231203682
https://journals.sagepub.com/doi/abs/10.1177/20563051231203682

18. The “boomerang effect”

Abelson, R.P., & Miller, J.C. (1967). Negative persuasion via personal insult. Journal of Experimental Social Psychology, 3(4), 321-333. https://doi.org/10.1016/0022-1031(67)90001-7
https://www.sciencedirect.com/science/article/abs/pii/0022103167900017

19. Hate-posters show elevated anger and anxiety after posting

Ghenai, A., Noorian, Z., Moradisani, H., Abadeh, P., Erentzen, C., & Zarrinkalam, F. (2025). Exploring hate speech dynamics: The emotional, linguistic, and thematic impact on social media users. Information Processing & Management, 62(3), 104079. https://doi.org/10.1016/j.ipm.2025.104079

https://doi.org/10.1016/j.ipm.2025.104079 (Note: the DOI in the article resolves to a related 2025 paper; the exact authorship of the specific cited study cannot not be fully verified from the locked page.)
https://www.sciencedirect.com/science/article/pii/S0306457325000214