Inclusive Intelligence-The Imperative of Diversity in AI.pptx

Learning objective: Explore how to address biases and discrimination embedded in AI systems that ensure responsible and ethical AI development and deployment.

In an era where artificial intelligence (AI) is becoming increasingly central to decision-making processes across industries, the imperative of diversity in AI development and deployment cannot be overstated. As AI algorithms rely heavily on data, they inherently reflect the biases and disparities prevalent in the datasets they are trained on. This issue underscores the importance of diversity not only in the workforce designing and deploying AI but also in the datasets used to train these systems. Without diverse representation and inclusive practices, AI technologies risk amplifying societal biases and perpetuating discrimination. This seminar delves into the fundamental necessity of diversity in AI, addressing the inherent biases and discrimination embedded in our societal fabric that AI systems risk perpetuating. By embracing diversity in AI development and governance, organizations can promote fairness, equity, and accountability in AI technologies, ultimately advancing towards a more just and inclusive future. Together, we can pave the way for AI technologies that reflect the diversity of our society and contribute positively to social progress and equality.

By the end of the seminar, attendees will:

1. Discuss the ethical implications of biased AI.
2. Examine strategies for mitigating algorithmic discrimination.
3. Explore the role of diversity and inclusion in AI development teams.

1. Inclusive Intelligence: The Imperative of Diversity in AI W omen of Color STEM Conference October 5, 2024 | Detroit, MI Moderator : Manvee Sharma, Client Partner - Infosys Panelists : Dr. Keysha A. Cutts, PMP & Stephanie Vaughn

2. In this talk, I will examine how efforts to move Artificial Intelligence forward have been built on statistical biases, and why the people and companies who build AI technology should care. I will also discuss what a code of ethics could look like when approached by developers. Abstract

3. Main Questions Do developers have an ethical responsibility when building tech that could potentially become a tool of state violence? Should developers have a say in how the tech they build is used, particularly after it has been delivered to the client?

4. Do developers have an ethical responsibility when it comes to building tech that could potentially become a tool of state violence?

5. “Acts of violence committed by an official state, military or sponsored by a sovereign government outside of the context of a declared war, which target civilians or show a disregard for civilian life in attacking targets—either people or facilities.” First, let’s define “state violence” — "State violence." Segen's Medical Dictionary

6. So what does this look like in practice? Two words: Facial Recognition .

7. A group of activists, organizers, and mathematicians committed to the mission of using data science to create concrete and measurable change in the lives of Black people. Data for Black Lives seeks to mobilize scientists around racial justice issues.

8. D4BL: Abolition in the Age of Big Data While most facial recognition systems are populated from passport photos or drivers license photos in the Federal database, many law enforcement databases (DBs) are populated with mug shots.

9. D4BL: Abolition in the Age of Big Data For areas with known issues around problematic policing, facial recognition DBs contain mostly mugshots of Black and brown faces.

10. D4BL: Abolition in the Age of Big Data When facial recognition DBs fail to operate, they don’t fail to identify - many misidentify .

11. D4BL Opening Panel: Abolition in the Age of Big Data Underneath it all, high arrest rates are being rewarded. The priority is to report numbers to the city, rather than maintaining a commitment to the well-being of the communities under their jurisdiction .

12. “Technology in the hands of those in power will continue to be used to oppress” — Data for Black Lives: Abolition in the Age of Big Data

13. Should developers have a say in how the tech they build is used once it’s delivered to the client?

14. “We call on Amazon to stop selling Rekognition to law enforcement as legislation and safeguards to prevent misuse are not in place.” — “ On Recent Research Auditing Commercial Facial Analysis Technology ” (March 26, 2019)

15. Caleb Thompson “Don’t Get Distracted” (RubyConf 2017) Sketchnotes illustration by Stephanie Nemeth

16. Conclusion: Why should we have a code of ethics in software development? How should we approach ethics in software development?

17. Why should we have a code of ethics in software development? “We spend the majority of our time working. Why would we want to put our life energy into something that consumes that much of our time without being thoughtful and intentional?” — Devney Hamilton, Sassafras Tech Collective

18. “Most professions have ethics standards and guidelines; software development is relatively new and that is taken for granted.” — Devney Hamilton, Sassafras Tech Collective Why should we have a code of ethics in software development?

19. “No one practicing software development should take the lack of a code of ethics for granted. It’s not needed any less. Software impacts our lives to the same degree as seeing a doctor or seeking counsel from a lawyer.” — Devney Hamilton, Sassafras Tech Collective Why should we have a code of ethics in software development?

20. Transparency; true ethics requires open source. Democratically controlled. Built by teams that are representative of the world we live in. — Devney Hamilton, Sassafras Tech Collective How should we approach ethics in software development?