My New Shrink, Claude
January 15, 2026
Sign up for blog updates!
Join my email list to receive updates and information.
I’ve used this Arthur C. Clarke quote in speeches for years: “Any sufficiently advanced technology is indistinguishable from magic.” This time I mean it.
I’ve become a daily devotee of an artificial intelligence (AI) app called Claude, and I can’t comprehend how it was devised by the hand of Man. It’s magic.
Claude was developed by Anthropic, an AI safety company founded in 2021 by former OpenAI members. The company focuses on building AI systems that are “safe, beneficial, and interpretable,” according to the company.
Claude was developed using both human and AI-generated feedback based on a set of principles known as Constitutional AI.
See why I can’t figure out how it works? It must be magic. And it’s free!
I began employing Claude some months ago to help me with research. Instead of visiting numerous websites to gather data, I had Claude do the roundup for me and present curated information to me. I run the same search using a different AI tool to make sure that what I’m getting is accurate.
But like Icarus, who rode his wax wings too close to the sun, I ventured into more esoteric uses of Claude. I started asking it (I was just now tempted to use the pronoun “him” rather than “it”) to gather information for my personal use.
For example, Claude devised a daily schedule for me that carved out time slots for work, recreation, meditation, and so forth.
Next I had him put together a daily meal plan to meet parameters I set for calories, proteins, carbs, and fiber.
Finally, I’ve been asking him to collate the counsel of psychologists. I asked Claude to recommend how I can better fulfill resolutions I make in the morning – but forget about by afternoon. Here’s some of his answer:
You might be setting yourself up with too much at once. When we're motivated in the morning, we tend to envision our best possible self doing everything. But willpower and decision-making energy are actually limited resources that deplete through the day. Starting with just one resolution - maybe the one that matters most to you - could be more sustainable than trying to overhaul everything simultaneously.
The "failure" might be information, not weakness. If you're consistently not following through, something about the resolution might not fit your actual life right now. Rather than seeing personal failure, you could treat it as feedback.
Consider starting absurdly small. The goal is to build the habit of keeping promises to yourself, even tiny ones.
Not bad advice. Although I can do without the contrived human touches, like “I really empathize with your frustration.”
And some of Claude’s ideas really did hit home:
- Morning-you and afternoon-you are almost different people. Morning-you is well-rested, hopeful, hasn't faced any difficulties yet. Afternoon-you is tired, has already made a hundred decisions, and that salad now feels like an obligation rather than something meaningful. The motivation gap isn't a character flaw. It's just that the person making the resolution isn't the same person who has to execute it.
- Motivation follows action more often than action follows motivation. This is counterintuitive, but waiting to "feel like it" often means waiting forever. The people who exercise regularly often don't feel particularly motivated in the moment – they just start.
Or as Nike’s ad agency so brilliantly put it, “Just do it.”
Hmm. I should ask Claude why Nancy Reagan’s “Just say no” campaign did a Thelma and Louise off a cliff.
Of course, like any wunderkind, Claude has limits.
I asked it to re-write this blog post in a “more professional, polished, and intelligent” style. Here’s Claude’s work product, which is why nobody should ever ask AI to write:
For years, I have invoked Arthur C. Clarke's observation in my presentations: "Any sufficiently advanced technology is indistinguishable from magic." Never has this axiom felt more apt than now.
I have become a dedicated user of an artificial intelligence application called Claude, and I genuinely struggle to comprehend the mechanisms underlying its capabilities. The technology borders on the inexplicable.
Developed by Anthropic—an AI safety company established in 2021 by former OpenAI executives—Claude represents a deliberate approach to building AI systems that prioritize safety, societal benefit, and interpretability. The platform was developed through Constitutional AI, a methodology incorporating both human and AI-generated feedback guided by explicit ethical principles.
The fact that I cannot fully grasp its operational framework only reinforces Clarke's point. And remarkably, it is available at no cost.
I initially adopted Claude several months ago as a research tool. Rather than consulting multiple websites to aggregate information, I now delegate this task to Claude, which synthesizes and presents curated data efficiently. I verify accuracy by cross-referencing results with alternative AI platforms.
However, like Icarus venturing too near the sun on waxen wings, I have gradually expanded my use of Claude into more personal domains. I find myself increasingly personifying the system—nearly referring to "it" as "him"—as I seek assistance with individual matters.
Claude has, for instance, constructed a daily schedule for me that allocates specific time blocks for professional work, recreation, meditation, and other activities. Subsequently, I requested a customized meal plan conforming to my specified parameters for calories, protein, carbohydrates, and fiber. Most recently, I have been consulting Claude to synthesize psychological research and therapeutic strategies. I posed a question about maintaining morning resolutions that tend to dissipate by afternoon. Claude's response included several noteworthy insights:
You may be attempting too much simultaneously. Morning optimism often generates unrealistic expectations of our capacity. However, willpower and decision-making ability are finite resources that diminish throughout the day. Beginning with a single resolution—perhaps the most consequential one—may prove more sustainable than attempting comprehensive behavioral change at once.Repeated non-compliance may constitute valuable data rather than evidence of weakness. Rather than interpreting this as personal failure, consider it diagnostic feedback about whether the resolution aligns with your current circumstances.Consider establishing absurdly modest initial goals. The objective is to develop the habit of honoring commitments to yourself, regardless of scale.
The counsel is sound, though I could do without the artificial empathy—phrases like "I really empathize with your frustration" feel contrived.
Nevertheless, certain observations resonated profoundly:
• Morning and afternoon versions of ourselves are fundamentally different entities. The morning self is rested, optimistic, and unburdened by the day's demands. The afternoon self is fatigued, depleted by countless decisions, and perceives that previously appealing salad as an obligation rather than a meaningful choice. This motivational discrepancy is not a character deficiency—it reflects the reality that the individual making the commitment differs from the one tasked with its execution.• Motivation more frequently follows action than precedes it. This counterintuitive principle suggests that waiting to "feel motivated" often means waiting indefinitely. Those who maintain consistent exercise routines rarely feel particularly motivated in the moment—they simply begin.
Or, as Nike's advertising memorably distilled it: "Just do it."
Which raises an intriguing question: I should ask Claude why Nancy Reagan's "Just Say No" campaign met such spectacular failure.