Ethical Challenges and Bias in Real-World AI: A Fairness-Oriented Approach to Resume Screening Systems
Bibliographic record
Abstract
Artificial Intelligence (AI) is rapidly integrating into decision-making across fields like finance, healthcare, and criminal justice. However, its widespread use brings ethical challenges, especially the potential for bias in AI algorithms. This paper examines these ethical concerns, emphasizing three main types of bias: data bias, algorithmic bias, and interaction biasThrough real-world case studies, including biased facial recognition systems, hiring algorithms, and criminal justice AI tools, this paper highlights the far-reaching societal impact of these biases. These case studies demonstrate how bias in AI systems can result in unequal access to resources, unjust treatment in legal systems, and perpetuation of stereotypes in employment practices. This paper highlights the pressing need for robust strategies to address the ethical risks associated with AI. It delves into established frameworks such as Ethically Aligned AI, the EU Guidelines on Trustworthy AI, and the Toronto Declaration, offering essential principles for creating AI systems that prioritize fairness, transparency, and accountability. By evaluating current applications across different industries, this paper proposes actionable solutions to reduce bias in AI systems and also a sample model has been developed to represent how the system has to be under ethics and including enhancing data diversity and implementing robust governance mechanisms to ensure ethical oversight in resume screening processThis research seeks to advance the conversation on ethical AI by emphasizing fairness, equity, and the social good, ensuring that AI technologies are developed to benefit all communities equally.
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 | 0.117 | 0.162 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.004 | 0.002 |
| Science and technology studies | 0.008 | 0.040 |
| Scholarly communication | 0.014 | 0.013 |
| Open science | 0.004 | 0.011 |
| Research integrity | 0.007 | 0.006 |
| Insufficient payload (model declined to judge) | 0.005 | 0.001 |
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".