AI Image Negative Prompt Ignored by Model
You added 'no text, no watermark, no extra fingers' and the output still has all three — negative prompt is either not wired up, too long, or fighting an overtrained concept.
#Prompt engineering
Articles tagged with #Prompt engineering
AI Video Camera Motion Goes Wrong Direction
You asked for a slow dolly-in and got a dolly-out. Or 'pan left' became 'pan right'. AI video models map motion vocabulary inconsistently — fix with explicit framing.
Few-Shot Examples Have Uneven Quality and Drag Output Down
You provided 5 few-shot examples. Two are great, three are mediocre. The model averages toward the mediocre ones. Why example quality variance hurts and how to curate.
Model Returns Invalid JSON Because Schema Was Described, Not Enforced
You asked for JSON matching a schema. Most calls return valid JSON, some return prose with JSON inside, some omit fields. Description vs enforcement, and how to fix at the API layer.
Model Invented Fake Citations and URLs
The model produced citations like Smith et al. 2019, journal of XYZ — and the paper does not exist. Or it linked to a URL that 404s. Why citation hallucination happens and how to stop it.
Model Replies in the Wrong Language
You prompted in English and the model answered in Chinese, or vice versa. Or it switched mid-output. Why language drift happens and how to lock the output language.
Response Got Cut Off Mid-Sentence Because max_tokens Was Too Low
The model's response ends abruptly in the middle of a sentence, a JSON object, or a code block. Almost always max_tokens. How to size it, detect truncation, and recover.
Prompt Asks for 10 Items, Model Returns 3 and Stops
You asked for 10 ideas, the model gave 3 and trailed off. Or it filled 10 slots but the last 4 are filler. Why list-length requests under-deliver and how to actually get N items.
Prompt Contains a Stale Year That Anchors the Model to Wrong Context
Your prompt template still says 2023 in 2026. Model anchors to 2023 context — old API versions, old pricing, old facts. Why date-staleness compounds and how to keep prompts evergreen.
AI Gives Lists When Execution Is Needed
You asked the model to do the work; it returned an outline of how someone could do the work.
AI Ignores the Most Important Constraint
You listed five rules. The model honored four and quietly dropped the one that actually mattered.
Ambiguous Evaluation Criteria Cause Weak Answers
You gave criteria, but the criteria are themselves vague — "engaging", "professional", "innovative" — so the model interprets freely.
Conflicting Instructions Weaken AI Output
Two parts of your prompt fight each other, so the model averages them and produces something nobody asked for.
Model Follows the Latest Sentence and Ignores the Earlier Rule
A casual aside at the end of your prompt overrides the careful rules you wrote at the top.
Missing Examples Leads to Output Drift
You described what you want in words; the model approximates. Add one concrete example and the approximation becomes a match.
Negative Constraints Are Too Vague
"Do not be generic" tells the model what not to do without telling it what to do.
No Output Format Specified — Why You Get a Wall of Prose
You wanted a structured answer; you got 600 words of paragraphs you cannot copy-paste anywhere.
Prompt Does Not State What Success Looks Like
Without a success criterion, "good" is whatever the model thinks looks confident.
Prompt Asks for "Best" Without Defining a Decision Rule
Without a decision rule, "best" defaults to whatever the model thinks sounds confident.
Prompt Copied From Another Task Fails Here
A prompt that worked great elsewhere produces nonsense for your current task because the assumptions do not transfer.
Prompt Uses Emotional Wording Instead of Operational Instructions
"Please make it amazing!" feels persuasive but tells the model nothing it can act on.
Prompt Lacks Context Hierarchy
You pasted everything as a flat block. The model cannot tell which lines are critical and which are background.
Prompt Lacks Source Material Hierarchy
You attached three documents and the model treated them as equally authoritative — including the outdated draft.
Prompt Does Not Define the Intended Audience
Without an audience, the model writes for an imaginary average reader and pleases no one.
Role Instruction Alone Is Not Enough
"You are a senior engineer" sets a vibe but does not change the deliverable. You still need rules, format, and examples.
Too Many Examples Overwhelm the Task
5+ examples can crowd out the instruction or push the model to imitate examples instead of executing the task.
Prompt Asks for Too Many Tasks at Once
You stacked five tasks in one prompt; the model did one well, one badly, and partially answered three.
AI Changes Scope Because the Task Boundary Is Unclear
You asked the model to do one thing; it also did three adjacent things you did not want.