AI Ethics Is a Double Misnomer
Why Read This
What Makes This Article Worth Your Time
Summary
What This Article Is About
Writing for Psychology Today, Dr. Cornelia C. Walther argues that the phrase “AI ethics” — now ubiquitous in boardrooms, policy papers, and university centres — contains two misleading compressions. The first is the word “artificial intelligence” itself: a metaphor that invites a category error by implying that current AI systems possess the living, embodied, aspirational, emotionally grounded quality of natural human intelligence. The second is “ethics,” which has been reduced to a governance checklist — fairness, transparency, explainability — when it is in fact humanity’s oldest inquiry into the good life, now operating inside recommendation engines and automated hiring tools.
Walther introduces her 4×4 framework of natural intelligence — four dimensions (aspirations, emotions, thoughts, sensations) operating across four levels (individual, community, country, planet) — to show how profoundly different human intelligence is from what AI simulates. Against this backdrop, she advocates for prosocial AI: an approach that embeds ethical reflection before systems are built and during their use, asking not merely whether an AI is fair but whether it deepens human agency, dignity, ecological responsibility, and the full texture of natural intelligence.
Key Points
Main Takeaways
“Intelligence” Is a Category Error
AI systems predict, classify, and generate, but lack embodiment, emotion, aspiration, and lived meaning — the foundations of natural intelligence. Calling them “intelligent” obscures this fundamental difference.
Ethics Predates Machine Learning
Moral questions — what do we owe one another, what powers require restraint — are ancient. AI does not create new ethics; it gives old questions a new operating system embedded in interfaces and automated decisions.
Natural Intelligence Is a 4×4 Living System
Human intelligence unfolds across four inner dimensions (aspirations, emotions, thoughts, sensations) and four collective levels (individual, community, country, planet) — all continuously shaping each other in ways AI cannot replicate.
Governance Frameworks Are Necessary but Insufficient
UNESCO, the OECD, and the EU AI Act all offer useful tools, but Walther argues they stop short of the deeper cultural question: do our AI systems actively cultivate and deepen natural intelligence in those who use them?
Ethical Reflection Belongs Before Design
Walther argues that ethical questions should enter AI development at the earliest stage — when purpose is still open and incentives are still negotiable — rather than being retrofitted as compliance requirements after a system is built.
Prosocial AI Is the Practical Path Forward
Prosocial AI asks whether a system strengthens human agency, dignity, inclusion, and ecological responsibility — turning moral aspiration into a measurable, correctable practice rather than a one-time compliance exercise.
Master Reading Comprehension
Practice with 365 curated articles and 2,400+ questions across 9 RC types.
Article Analysis
Breaking Down the Elements
Main Idea
Our Language About AI Is Shaping — and Limiting — Our Moral Imagination
Walther argues that the very words “AI ethics” prevent us from asking the right questions. By calling computational systems “intelligent,” we anthropomorphise them and misplace moral responsibility. By calling governance checklists “ethics,” we trivialise a 3,000-year-old inquiry into what it means to live well. The remedy is not new regulation but a richer, more honest vocabulary and a deeper set of questions about human becoming.
Purpose
To Reframe the AI Ethics Debate Around Human Flourishing, Not Compliance
Walther’s purpose is philosophical and corrective: she wants to shift the conversation from “are our AI systems safe and fair?” — a compliance question — to “who are we becoming with these systems?” — a humanistic one. Writing in Psychology Today gives her argument reach beyond the technical and policy communities, targeting readers who live with AI’s daily effects on attention, relationships, and identity.
Structure
Critique → Framework → Governance Survey → Prescriptive Vision
The article opens by naming the double misnomer, unpacks each word in turn, introduces the 4×4 natural intelligence framework as the positive contrast, surveys existing governance responses and their limitations, then closes with the prosocial AI proposal. This Analytical → Expository → Critical → Prescriptive arc is disciplined and cumulative — each section earns the next, ending with a call to action rather than mere critique.
Tone
Measured, Philosophical & Quietly Urgent
Walther writes with scholarly precision but genuine moral seriousness — the tone is neither alarmist nor dismissive of AI’s value. There is a quiet urgency throughout, conveyed not through dramatic language but through the weight of her questions: “Who are we becoming with these systems?” The piece reads less like a policy critique and more like a philosophical intervention aimed at expanding what we think is at stake.
Key Terms
Vocabulary from the Article
Click each card to reveal the definition
Build your vocabulary systematically
Each article in our course includes 8-12 vocabulary words with contextual usage.
Tough Words
Challenging Vocabulary
Tap each card to flip and see the definition
To attribute human characteristics, emotions, or intentions to non-human entities, potentially distorting understanding of their actual nature and capabilities.
“They can produce language that sounds reflective… The word intelligence invites a category error.”
The degree to which an AI system’s decisions and outputs can be understood and accounted for by humans, especially when those decisions affect people’s lives or rights.
“Fairness, transparency, explainability, and accountability matter. They form part of the necessary infrastructure.”
The process of sorting patients, cases, or priorities according to urgency and available resources — used here to indicate AI’s expanding role in high-stakes medical decision-making.
“It places them inside recommendation engines, automated hiring tools, clinical triage, personalized learning…”
A confusing or difficult problem or question with no clear or easy solution, especially one with competing considerations that resist simple resolution.
“The C of the climate conundrum is one issue among a whole ABCD of underappreciated AI-issues.”
The process by which organisations obtain goods, services, or systems — used here to show how “AI ethics” has become an official requirement built into purchasing and contracting rules.
“It appears in boardrooms, policy papers, university centers, procurement rules, and product reviews.”
The condition of mutual reliance between entities — people, communities, species, systems — where each depends on the health and functioning of the others for its own wellbeing.
“Do they help people think with care, feel with maturity, act with responsibility, and sense their interdependence with others and the planet?”
Reading Comprehension
Test Your Understanding
5 questions covering different RC question types
1According to Walther, the phrase “AI ethics” is problematic primarily because it is too new a term to have developed a clear meaning and lacks sufficient academic research behind it.
2According to the article, what statistic from Stanford’s 2026 AI Index does Walther cite to show that AI is becoming “part of the cognitive atmosphere”?
3Which sentence best captures Walther’s argument for why existing “AI ethics” governance frameworks — like those from UNESCO, the OECD, and the EU — are not enough?
4Evaluate each of the following statements about Walther’s 4×4 framework of natural intelligence:
The four inner dimensions of natural intelligence identified by Walther are aspirations, emotions, thoughts, and sensations.
Walther uses the example of a polluted city to illustrate how collective-level conditions (the planet) can affect individual-level dimensions (lungs and moods).
According to the article, the four collective levels of the 4×4 framework are individual, family, community, and country.
Select True or False for all three statements, then click “Check Answers”
5When Walther states that “every system teaches something through repeated use,” what can most reasonably be inferred about her view of AI’s relationship to human nature?
FAQ
Frequently Asked Questions
Walther argues that AI systems can reproduce certain outputs of natural intelligence — language generation, pattern recognition, classification — but only the measurable, computable outputs. They cannot simulate the living process behind those outputs: the embodied experience, emotional grounding, aspiration, social belonging, and moral development that make human intelligence what it is. The simulation can be useful, even extraordinary, but it remains fundamentally incomplete.
Published in 2021, “On the Dangers of Stochastic Parrots” warned about the environmental costs of training large language models, problems with dataset documentation, encoded bias, and crucially the risk of mistaking fluent language production for genuine understanding. Walther references it to support her first misnomer argument — that the appearance of intelligence in AI output does not constitute intelligence in any meaningful sense — and notes this warning has grown more urgent as models have become more fluent.
Standard AI ethics frameworks — fairness, transparency, explainability, accountability — ask whether a system is compliant and safe. Prosocial AI asks a deeper question: does this system strengthen or erode the dimensions of natural intelligence in the people who use it? It shifts ethical reflection upstream into the design phase, before incentives are fixed, and downstream into everyday use, monitoring whether real human capacities — aspiration, emotional maturity, independent thought, ecological awareness — are being cultivated or diminished.
Readlite provides curated articles with comprehensive analysis including summaries, key points, vocabulary building, and practice questions across 9 different RC question types. Our Ultimate Reading Course offers 365 articles with 2,400+ questions to systematically improve your reading comprehension skills.
This article is rated Advanced. While written accessibly for a Psychology Today audience, it engages with philosophical distinctions — category errors, the nature of simulation versus reality, the difference between compliance and genuine ethics — that require careful, active reading. The argument is layered and cumulative, and key claims are compressed into single sentences that reward re-reading. Readers must also track a dual-thread critique (two misnomers) and evaluate two different levels of governance response.
Cornelia C. Walther is a Ph.D.-level researcher and writer who publishes on hybrid intelligence, AI governance, and human development in Psychology Today. Her work draws on psychology, development theory, and ethics to examine how AI systems affect human capacities and social wellbeing. Her 4×4 framework of natural intelligence reflects a cross-disciplinary approach that situates AI questions within a broader understanding of what human flourishing actually requires.
The Ultimate Reading Course covers 9 RC question types: Multiple Choice, True/False, Multi-Statement T/F, Text Highlight, Fill in the Blanks, Matching, Sequencing, Error Spotting, and Short Answer. This comprehensive coverage prepares you for any reading comprehension format you might encounter.
