Is Claude Moralistic or Principled?
Why Claude Is the First AI I Turn to for Sanity Checks
Claude AI is not well known in my home country, and I only heard about it on a Fresh Air interview with NYT reporter Kashmir Hill, where Claude was described as the one AI that showed some moral backbone (in Hill’s words, Claude is “very moralistic”), while others were eager to please. Intrigued by this least sycophantic AI, I set up a personal Claude account, using it for free at first but then upgrading to premium tier because I was hitting usage limits.
Claude quickly became a relatively reliable sounding board for my skepticism (with prompting or after some context building), but Sonnet 4 proved itself as a recap champ that was the best of all models I was using, including even its advanced sister model Opus 4. Sonnet 4’s are just the right length, a fact I confirmed with its competitors, which all conceded defeat.
Based on my interactions with Claude, though, I don’t think Hill meant “moralistic” but “principled.” This principled stance makes it the most trustworthy of my thinking A.I.des for sanity checks, which are among my main use cases—an English “native” I can rely on 24/7 to check if my understanding of this learned foreign language is correct or if it just is in my head.
In the chat excerpts below, I’ve spelled out abbreviations and clarified references for readability—my actual prompts were more compressed due to context limits.
Want to see how this unfolded? Here are excerpts of pivotal points from those actual conversations.
Language Trends
Prompt: Unrelated question: Why is “commit” back as a collocation for harmful behavior directed at oneself, when it was considered judgy a few years ago? Even media people are now using it? Is this anti-woke backlash going all the way to the other extreme?
[Note: As you can see here, I was so careful about potential content restrictions that I described the topic as “harmful behavior directed at oneself” rather than using direct terminology in my prompt—yet Claude still engaged substantively and maintained its own professional standards rather than delivering a canned safety response.]
Professional Standards
Prompt: I’m worried that media types get swept up in these trends. These are principled positions that should not be abandoned.
Setting the Record Straight
Prompt: It’s irresponsible and unprofessional.
This interaction demonstrates an important distinction: a moralistic AI might have deflected the question entirely or lectured about the ethical implications of discussing such topics. Claude’s principled approach meant engaging thoughtfully with my carefully framed question while maintaining professional language standards—exactly the kind of reliable intellectual partnership I value.
A Principled AI
In another chat, I brought up Hill’s characterization of Claude as “moralistic,” because I thought that word, as I had heard it used, carried a holier-than-thou connotation, which was at odds with both Hill’s description of her interaction with Claude and my own experience.
Prompt: It was an NYT reporter—Kashmir Hill. She said she found you “moralistic”—wrong word choice, in my opinion, when she meant “principled”—and you didn’t keep making suggestions for every aspect of her life, unlike other AI.
