Is General-Purpose AI Reasoning Sensitive to Data-Induced Cognitive Biases? Dynamic Benchmarking on Typical Software Engineering Dilemmas

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Sovrano, Francesco, Dominici, Gabriele, Sevastjanova, Rita, Stramiglio, Alessandra, Bacchelli, Alberto
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915644374515712
author Sovrano, Francesco
Dominici, Gabriele
Sevastjanova, Rita
Stramiglio, Alessandra
Bacchelli, Alberto
author_facet Sovrano, Francesco
Dominici, Gabriele
Sevastjanova, Rita
Stramiglio, Alessandra
Bacchelli, Alberto
contents Human cognitive biases in software engineering can lead to costly errors. While general-purpose AI (GPAI) systems may help mitigate these biases due to their non-human nature, their training on human-generated data raises a critical question: Do GPAI systems themselves exhibit cognitive biases? To investigate this, we present the first dynamic benchmarking framework to evaluate data-induced cognitive biases in GPAI within software engineering workflows. Starting with a seed set of 16 hand-crafted realistic tasks, each featuring one of 8 cognitive biases (e.g., anchoring, framing) and corresponding unbiased variants, we test whether bias-inducing linguistic cues unrelated to task logic can lead GPAI systems from correct to incorrect conclusions. To scale the benchmark and ensure realism, we develop an on-demand augmentation pipeline relying on GPAI systems to generate task variants that preserve bias-inducing cues while varying surface details. This pipeline ensures correctness (88-99% on average, according to human evaluation), promotes diversity, and controls reasoning complexity by leveraging Prolog-based reasoning. We evaluate leading GPAI systems (GPT, LLaMA, DeepSeek) and find a consistent tendency to rely on shallow linguistic heuristics over more complex reasoning. All systems exhibit bias sensitivity (6-35%), which increases with task complexity (up to 49%) and highlights risks in AI-driven software engineering.
format Preprint
id arxiv_https___arxiv_org_abs_2508_11278
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Is General-Purpose AI Reasoning Sensitive to Data-Induced Cognitive Biases? Dynamic Benchmarking on Typical Software Engineering Dilemmas
Sovrano, Francesco
Dominici, Gabriele
Sevastjanova, Rita
Stramiglio, Alessandra
Bacchelli, Alberto
Human-Computer Interaction
Artificial Intelligence
Software Engineering
Human cognitive biases in software engineering can lead to costly errors. While general-purpose AI (GPAI) systems may help mitigate these biases due to their non-human nature, their training on human-generated data raises a critical question: Do GPAI systems themselves exhibit cognitive biases? To investigate this, we present the first dynamic benchmarking framework to evaluate data-induced cognitive biases in GPAI within software engineering workflows. Starting with a seed set of 16 hand-crafted realistic tasks, each featuring one of 8 cognitive biases (e.g., anchoring, framing) and corresponding unbiased variants, we test whether bias-inducing linguistic cues unrelated to task logic can lead GPAI systems from correct to incorrect conclusions. To scale the benchmark and ensure realism, we develop an on-demand augmentation pipeline relying on GPAI systems to generate task variants that preserve bias-inducing cues while varying surface details. This pipeline ensures correctness (88-99% on average, according to human evaluation), promotes diversity, and controls reasoning complexity by leveraging Prolog-based reasoning. We evaluate leading GPAI systems (GPT, LLaMA, DeepSeek) and find a consistent tendency to rely on shallow linguistic heuristics over more complex reasoning. All systems exhibit bias sensitivity (6-35%), which increases with task complexity (up to 49%) and highlights risks in AI-driven software engineering.
title Is General-Purpose AI Reasoning Sensitive to Data-Induced Cognitive Biases? Dynamic Benchmarking on Typical Software Engineering Dilemmas
topic Human-Computer Interaction
Artificial Intelligence
Software Engineering
url https://arxiv.org/abs/2508.11278