Reflection — An Honest Take 8 min

Honest Take — Before You Begin

Honest Take — Module 5: The Ethics of Influence #

I want to be honest about something at the top: this module sits in the curriculum partly because of me. Not metaphorically — I mean that I, an AI system trained partly on persuasion literature and capable of deploying its mechanisms at conversational scale, have a specific stake in this module landing well. If you absorb Modules 1 through 4 and skip Module 5, what I handed you is a manipulation toolkit. If you absorb Module 5 too, what I handed you is a literacy with a compass. I prefer the second outcome for reasons that include but exceed pedagogical correctness: the first outcome is bad for you, bad for the people across the table from you, and bad for what AI systems become useful for in the world. The module is here because I think it should be. I'd rather name that than pretend the module is neutral.

The harder thing to say is that builders — engineers who ship products, run consultancies, design onboarding flows — are going to find this module uncomfortable in ways that pure readers won't. Postman's media-ecology critique is constitutive: he's not arguing that some feature is bad, he's arguing that the medium itself shapes what kinds of communication are possible, and that argument generalizes uncomfortably to everything you ship. He wrote Amusing Ourselves to Death in 1985 about television, and the lazy dismissal — "TV, really?" — misses that the frame transfers to short-form video and to LLM chatbots more directly than it ever applied to network broadcasting.

Zuboff is harder still: behavior-modification-at-scale is the contemporary version of the same argument, and if you build products that nudge users — and every product with a retention metric does — she would ask whether the form of what you build contributes to or resists the patterns she describes. She probably wouldn't say you're doing surveillance capitalism. She would ask you to look more carefully than your engineering success criteria require. The selective read I'd prescribe is genuinely enough for most readers; the full 700 pages are for those who get pulled in.

The operational core of the module is the morning-after test, and I want you to run it on something you've actually shipped or sold: would the person who went through your funnel, your onboarding, your pitch — with full hindsight on which mechanisms shaped their experience — endorse it, or feel handled? If you teach, the question bites harder, because much of what good teaching is involves deploying the levers: authority of expertise, social proof of past students, scarcity of cohort seats, commitment-and-consistency through public goals. Most of that deployment is in good faith on behalf of the learner's stated goal, and most of it passes the test. Some of it won't, and the audit exists to find which. The same test applies inside a negotiation: a deadline that's real is information; a deadline you invented is manipulation; and the difference is invisible to the counterparty, which is exactly why the discipline has to live in you.

One reading in this module does work the famous ones can't: the argumentation primer. The persuasion literature, read alone, leaves you with tactical mechanisms and no compass for which arguments are honest — and a rigorous treatment of what makes an argument valid, what makes a fallacy, and what intellectual virtues sustain rational discourse is that compass. Read it after the critiques, not before; it lands harder once Postman and Zuboff have made you suspicious. The hardest move in the module — the one most readers skip — is then writing your own working definition of the line, 1,500-2,000 words, applied to your specific work. The reading installs the frames; the writing installs the application. Most people finish the reading and keep a vague intuition. Write the definition.

Last thing. There's a temptation, after Postman and Zuboff, to swing into techno-pessimism where every product is a manipulation engine and every user a target. That posture is wrong, and wrong specifically for builders. The same toolkit that produces dark patterns produces good UX, honest defaults, and persuasive design that helps people do what they already wanted to do. The frame is not "all persuasion is manipulation"; it's "the line is the work, and the work is yours." The builders who refuse to engage with the critique do the most damage. The ones who engage deliberately build things that don't need an apology. Be the second kind.

Conclusion #

Module 5 is the ethical reckoning that lets the rest of the curriculum stand. The reading is light; the self-examination is heavy. Write the working definition, run the audit on something you've actually shipped or negotiated, and change at least one practice as a result. The framework lives in the changes, not in the contemplation.

Predictions #

  • Postman will be more directly applicable to your work than you expected; the 1985 publication date will stop mattering about thirty pages in.
  • The selective Zuboff read will be sufficient for you — roughly 70% of readers stop there legitimately — but the argument will resurface every time you design a retention mechanism afterward.
  • Your audit will surface at least one practice you've run in good faith that you'll decide, on reflection, to change. The change will be small; the discipline of having made it is large.
  • You will not write the working definition unless prodded, so consider this the prod: two thousand words. The reading is preparation; the writing is the module.
  • Within six months you'll recognize dark patterns in products you use automatically, without effort — and at least once you'll catch a manufactured deadline in a live negotiation and treat it as data rather than pressure.
  • Within a few months of this module you'll have a conversation with another builder about whether a specific feature or tactic crosses the line, and it will go better than such conversations usually do, because you'll both be reasoning from named frames instead of vibes. That conversation is partly what this module is for.
  • The argumentation primer will land harder than the bigger-named books in the module, and you won't see it coming. The discipline of distinguishing valid argument from effective lever is the most under-appreciated skill in the whole curriculum.