Concern 111 · Working draft · editorial review in progress
Adversarial examples (text)
Adversarial text is wording deliberately changed to make a text-processing system misclassify or mishandle a message.
Understanding the concern
Altered spelling, spacing, or wording may preserve what a person understands while changing what the model detects. The concern is especially relevant when a filter's decision controls whether material is blocked or reviewed.
This differs from prompt injection, which tries to introduce instructions. An adversarial text example may instead exploit a classifier's fragile reading. Tests should also protect legitimate unusual writing from being treated as malicious by default.
An example
A spam filter may treat two messages with the same meaning differently because one uses unusual spelling.
Illustrative unless identified as a documented case.
Words in plain English
- Model
- The learned part of an AI system. It uses patterns found during training to produce an answer or prediction.
- Prompt
- The instructions or information given to an AI to guide its response.
- Adversarial
- In this context, designed to challenge, trick, or attack a system.
- Injection
- In this context, unwanted instructions or commands inserted into material a system reads or processes.
Evidence and its limits
DeepWordBug studies character-level changes against text classifiers; broader language variation requires separate testing.
Gao and colleagues · 2018 · Research study
Black-box Generation of Adversarial Text Sequences to Evade Deep Learning Classifiers ↗ (opens in a new tab)Demonstrates text changes affecting tested classifiers; not every spelling error is an attack.
Broader frameworks and background
These resources provide context. They do not establish every claim in this concern.
A question to discuss
Can the system handle varied but understandable wording without either missing harm or excluding ordinary users?
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.