# FATE in AI: Towards Algorithmic Inclusivity and Accessibility

**Authors:** Isa Inuwa-Dutse  
**arXiv:** [2301.01590](https://arxiv.org/abs/2301.01590)  
**Date:** 2023-01-03  
**Summary page:** https://www.summarizepaper.com/en/paper/2301.01590v1/

## Key points

- The widespread deployment of AI technology is affecting many sections of society with economic and personal implications.
- Issues of fairness, accountability, transparency, and ethics (FATE) have arisen to address algorithmic disparities in AI.
- Ethical AI aims to incorporate moral behavior to avoid encoding bias in AI's decisions.
- Concerns have been raised regarding neglecting local knowledge, cultural pluralism, and global fairness in the present discourse on critical issues shaped by more economically developed countries (MEDC).
- A study was conducted to assess FATE-related issues and the effectiveness of transparency methods with a focus on underserved areas in the Global South.
- An online user survey was conducted with 43 participants to collect relevant data and assess FATE-related issues using ads data from online social networks.
- Explanations about decisions reached by AI systems tend to be vague and less informative.
- There is a need to engage with the community for the best way to integrate FATE into decision-making processes to bridge the accessibility and inclusivity gap in AI.
- For a successful national AI strategy in Africa where organizations offer AI-supported services across some countries, there is a need for algorithmic accountability, data protection, and explainability of decision-making by AI models.
- It is crucial not only to incorporate sociodemographic factors but also viewpoints from affected communities when deploying AI systems globally.

## Summary

The widespread deployment of systems powered by artificial intelligence (AI) technology has become a defining phenomenon in this age. Algorithmic decisions made by AI are affecting many sections of society, both directly and indirectly, with economic and personal implications. The issues of fairness, accountability, transparency, and ethics (FATE) have arisen to address algorithmic disparities in AI. Ethical AI aims to incorporate moral behavior to avoid encoding bias in AI's decisions. However, the present discourse on such critical issues is being shaped by more economically developed countries (MEDC), which raises concerns regarding neglecting local knowledge, cultural pluralism, and global fairness. To address these concerns, a study was conducted that builds upon existing research on responsible AI with a focus on areas in the Global South considered underserved vis-a-vis AI. The study's goal is twofold: firstly, to assess FATE-related issues and the effectiveness of transparency methods; secondly, to proffer useful insights and stimulate action towards bridging the accessibility and inclusivity gap in AI. The approach involved both qualitative and quantitative methods. An online user survey was conducted with 43 participants to collect relevant data and assess FATE-related issues using ads data from online social networks. The study found that explanations about decisions reached by the AI systems tend to be vague and less informative. To bridge the accessibility and inclusivity gap in AI there is a need to engage with the community for the best way to integrate fairness, accountability, transparency, and ethics into their decision making processes. For a successful national AI strategy in Africa where organisations such as African Development Bank offer AI-supported services across some countries [59], there is a need to ensure algorithmic accountability, data protection, and explainability of decision-making by AI models. In conclusion, as Artificial Intelligence systems continue to be deployed across various domains globally [7-14], it is crucial not only to incorporate sociodemographic factors but also viewpoints from affected communities to avoid promoting bias or unfair algorithmic decisions. This study presents a useful approach for examining FATE related issues with an emphasis on areas not traditionally served by Artificial Intelligence systems.