The formal rationality of artificial intelligence-based algorithms and the problem of bias
Bibliographic record
Abstract
This paper presents a new perspective on the problem of bias in artificial intelligence (AI)-driven decision-making by examining the fundamental difference between AI and human rationality in making sense of data. Current research has focused primarily on software engineers’ bounded rationality and bias in the data fed to algorithms but has neglected the crucial role of algorithmic rationality in producing bias. Using a Weberian distinction between formal and substantive rationality, we inquire why AI-based algorithms lack the ability to display common sense in data interpretation, leading to flawed decisions. We first conduct a rigorous text analysis to uncover and exemplify contextual nuances within the sampled data. We then combine unsupervised and supervised learning, revealing that algorithmic decision-making characterizes and judges data categories mechanically as it operates through the formal rationality of mathematical optimization procedures. Next, using an AI tool, we demonstrate how formal rationality embedded in AI-based algorithms limits its capacity to perform adequately in complex contexts, thus leading to bias and poor decisions. Finally, we delineate the boundary conditions and limitations of leveraging formal rationality to automatize algorithmic decision-making. Our study provides a deeper understanding of the rationality-based causes of AI’s role in bias and poor decisions, even when data is generated in a largely bias-free context.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.065 | 0.169 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.002 |
| Bibliometrics | 0.004 | 0.002 |
| Science and technology studies | 0.003 | 0.045 |
| Scholarly communication | 0.008 | 0.013 |
| Open science | 0.003 | 0.006 |
| Research integrity | 0.004 | 0.005 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".