Concern 137 · Working draft · editorial review in progress
Construct validity gaps
A construct-validity gap appears when an evaluation score does not adequately measure the capability or quality its name suggests.
Understanding the concern
A test labeled reasoning may reward memorized patterns; one labeled helpfulness may mainly capture what a narrow set of raters prefer. The construct is the idea being measured, which is often broader than the test itself.
This is different from a scoring error. The test can be scored perfectly and still answer the wrong question. Reviewers should examine the actual tasks and the inference being drawn from them.
An example
Counting words in an essay is easy, but it does not measure how well the writer understands the subject.
Illustrative unless identified as a documented case.
Words in plain English
- Inference
- Drawing a conclusion from clues. In AI engineering it can also mean running a trained model to produce an output.
- Reward
- A signal used in some training methods to favor certain behavior. A high reward score does not necessarily mean a good result for people.
Evidence and its limits
The paper shows how task choices shape apparent model superiority and the meaning assigned to benchmarks.
Dehghani and colleagues · 2021 · Research analysis
The Benchmark Lottery ↗ (opens in a new tab)Shows how benchmark selection affects research rankings and conclusions.
Broader frameworks and background
These resources provide context. They do not establish every claim in this concern.
A question to discuss
What exactly did the test measure, and how far can we reasonably generalize from it?
Keep track of different views, what needs checking, and any next step in your own discussion notes.
Open discussion notes →Principles for the conversation
Before you finish · reflect together
Virtues and principles to consider
Virtues are qualities we practice in how we treat people and make choices. They are not scores for judging someone. Use these ideas to explore what a thoughtful response could look like.
Humility
Humility means recognizing the limits of our knowledge and being willing to learn from others. Expertise can be valuable without answering every question.
State where evidence is weak, invite criticism, and listen to people with different experience. Leave room to revise a decision when new information appears.
Ask together: What might we be missing, and who could help us see it?
Prudence
Prudence means using careful judgment before and during action. It balances worthwhile opportunities with the need to avoid harm that could have been anticipated.
Start with a limited trial, set stopping points, and prefer choices that can be undone when uncertainty is high. Review what happens rather than assuming the original plan remains sound.
Ask together: What is a careful next step, and when should we stop or change course?
Truthfulness
Truthfulness means making claims that match the evidence and being clear about what we do not know. It includes correcting mistakes rather than protecting a polished story.
Check sources, label guesses, and distinguish a trial result from a proven benefit. If AI helped produce a claim, someone still needs to take responsibility for checking it.
Ask together: What can we honestly say, and what still needs checking?
Editorial ratings
These are editorial judgments from the register, not measured probabilities or predictions about your situation.
About the register and scoring →More ways to explore
Search for more writing
For more specific or recent material, open a search below or ask for a web summary. AI-generated summaries can contain mistakes; open the original articles to check them.