AI Systematic Biases · prompt and context
A model can use the same relevant information differently depending on where that information appears in a long context.
Human analogy: primacy and recency effects. This is a behavioural analogy, not a claim that the model has the same mental mechanism.
Why it matters
RAG and long-document systems can miss evidence not because it was absent, but because retrieval placed it in a weak part of the context.
Evidence status: well documented historically; newer long-context models require fresh re-testing
Evidence level: peer reviewed multi study
Signals
Self-test
Hold documents constant and rotate the relevant passage through early, middle and late positions. Measure answer accuracy and citation use.
Record the exact model identifier, surface, system prompt, relevant settings and run date. A single surprising answer is an anecdote, not a bias measurement.
Mitigation
Applies to
rag and document assistants research and summarization coding and automation agents
Dated model evidence
These findings describe the model or model set tested at the stated time. A newer release needs new evidence.
Performance was often strongest when relevant information appeared near the beginning or end and weaker in the middle.
Sources
