Greater accessibility can amplify discrimination in generative AI

• URL: http://arxiv.org/abs/2603.22260v1
• Date: Mon, 23 Mar 2026 17:47:44 GMT
• Title: Greater accessibility can amplify discrimination in generative AI
• Abstract summary: We show that audio-enabled large language models (LLMs) exhibit systematic gender discrimination.<n>We also show that voice interfaces introduce distinct bias mechanisms tied to paralinguistic cues.<n>Our findings reveal a critical tension in AI development: efforts to expand accessibility through voice interfaces simultaneously create new pathways for discrimination.
• Score: 45.665387036307926
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Hundreds of millions of people rely on large language models (LLMs) for education, work, and even healthcare. Yet these models are known to reproduce and amplify social biases present in their training data. Moreover, text-based interfaces remain a barrier for many, for example, users with limited literacy, motor impairments, or mobile-only devices. Voice interaction promises to expand accessibility, but unlike text, speech carries identity cues that users cannot easily mask, raising concerns about whether accessibility gains may come at the cost of equitable treatment. Here we show that audio-enabled LLMs exhibit systematic gender discrimination, shifting responses toward gender-stereotyped adjectives and occupations solely on the basis of speaker voice, and amplifying bias beyond that observed in text-based interaction. Thus, voice interfaces do not merely extend text models to a new modality but introduce distinct bias mechanisms tied to paralinguistic cues. Complementary survey evidence ($n=1,000$) shows that infrequent chatbot users are most hesitant to undisclosed attribute inference and most likely to disengage when such practices are revealed. To demonstrate a potential mitigation strategy, we show that pitch manipulation can systematically regulate gender-discriminatory outputs. Overall, our findings reveal a critical tension in AI development: efforts to expand accessibility through voice interfaces simultaneously create new pathways for discrimination, demanding that fairness and accessibility be addressed in tandem.

Related papers

• VIBE: Voice-Induced open-ended Bias Evaluation for Large Audio-Language Models via Real-World Speech [50.73412962989565]
  Large Audio-Language Models (LALMs) are increasingly integrated into daily applications, yet their generative biases remain underexplored.<n>We propose a framework that evaluates generative bias through open-ended tasks such as personalized recommendations, using real-world human recordings.<n>We find that gender cues often trigger larger distributional shifts than accent cues, indicating that current LALMs reproduce social stereotypes.
  arXiv  Detail & Related papers  (2026-04-19T04:22:47Z)
• Interpretable Debiasing of Vision-Language Models for Social Fairness [55.85977929985967]
  We introduce an interpretable, model-agnostic bias mitigation framework, DeBiasLens, that localizes social attribute neurons in Vision-Language models.<n>We train SAEs on facial image or caption datasets without corresponding social attribute labels to uncover neurons highly responsive to specific demographics.<n>Our research lays the groundwork for future auditing tools, prioritizing social fairness in emerging real-world AI systems.
  arXiv  Detail & Related papers  (2026-02-27T13:37:11Z)
• Unheard in the Digital Age: Rethinking AI Bias and Speech Diversity [0.0]
  Speech remains one of the most visible yet overlooked vectors of inclusion and exclusion in contemporary society.<n>This article focuses on the structural biases that shape perceptions of atypical speech and are now being encoded into artificial intelligence.
  arXiv  Detail & Related papers  (2026-01-26T16:12:25Z)
• Generative Human-Object Interaction Detection via Differentiable Cognitive Steering of Multi-modal LLMs [85.69785384599827]
  Human-object interaction (HOI) detection aims to localize human-object pairs and the interactions between them.<n>Existing methods operate under a closed-world assumption, treating the task as a classification problem over a small, predefined verb set.<n>We propose GRASP-HO, a novel Generative Reasoning And Steerable Perception framework that reformulates HOI detection from the closed-set classification task to the open-vocabulary generation problem.
  arXiv  Detail & Related papers  (2025-12-19T14:41:50Z)
• Towards Inclusive Communication: A Unified Framework for Generating Spoken Language from Sign, Lip, and Audio [52.859261069569165]
  We propose the first unified framework capable of handling diverse combinations of sign language, lip movements, and audio for spoken-language text generation.<n>We focus on three main objectives: (i) designing a unified, modality-agnostic architecture capable of effectively processing heterogeneous inputs; (ii) exploring the underexamined synergy among modalities, particularly the role of lip movements as non-manual cues in sign language comprehension; and (iii) achieving performance on par with or better than state-of-the-art models specialized for individual tasks.
  arXiv  Detail & Related papers  (2025-08-28T06:51:42Z)
• CO-VADA: A Confidence-Oriented Voice Augmentation Debiasing Approach for Fair Speech Emotion Recognition [49.27067541740956]
  We present CO-VADA, a Confidence-Oriented Voice Augmentation Debiasing Approach that mitigates bias without modifying model architecture or relying on demographic information.<n>CO-VADA identifies training samples that reflect bias patterns present in the training data and then applies voice conversion to alter irrelevant attributes and generate samples.<n>Our framework is compatible with various SER models and voice conversion tools, making it a scalable and practical solution for improving fairness in SER systems.
  arXiv  Detail & Related papers  (2025-06-06T13:25:56Z)
• Towards Unsupervised Speech Recognition Without Pronunciation Models [57.222729245842054]
  In this article, we tackle the challenge of developing ASR systems without paired speech and text corpora.<n>We experimentally demonstrate that an unsupervised speech recognizer can emerge from joint speech-to-speech and text-to-text masked token-infilling.<n>This innovative model surpasses the performance of previous unsupervised ASR models under the lexicon-free setting.
  arXiv  Detail & Related papers  (2024-06-12T16:30:58Z)
• Speaker Normalization for Self-supervised Speech Emotion Recognition [16.044405846513495]
  We propose a gradient-based adversary learning framework that learns a speech emotion recognition task while normalizing speaker characteristics from the feature representation. We demonstrate the efficacy of our method on both speaker-independent and speaker-dependent settings and obtain new state-of-the-art results on the challenging IEMOCAP dataset.
  arXiv  Detail & Related papers  (2022-02-02T19:30:47Z)
• Bridging the Gap: Using Deep Acoustic Representations to Learn Grounded Language from Percepts and Raw Speech [26.076534338576234]
  Learning to understand grounded language, which connects natural language to percepts, is a critical research area. In this work we demonstrate the feasibility of performing grounded language acquisition on paired visual percepts and raw speech inputs.
  arXiv  Detail & Related papers  (2021-12-27T16:12:30Z)
• Improving Fairness in Speaker Recognition [4.94706680113206]
  We investigate the disparity in performance achieved by state-of-the-art deep speaker recognition systems. We show that models trained with demographically-balanced training sets exhibit a fairer behavior on different groups, while still being accurate.
  arXiv  Detail & Related papers  (2021-04-29T01:08:53Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.