Leveraging Artificial Intelligence to Fight Against Discrimination: A Personal Example
AI and Inequality: How Machines Keep Us Poor, Sick, and Discriminated Against
Attending the AI and Inequality conference was a profoundly impactful experience for me. As a displaced tech worker and an advocate for ethical AI through my nonprofit organization, Voice For Change, I understand the transformative potential of AI technologies. This event, curated and moderated by Dr. Avriel Epps, featured prominent speakers like Meredith Broussard, Virginia Eubanks, and Charles Senteio. In this third and final panel, they discussed the inherent biases in AI technologies, particularly as they are deployed in law enforcement, healthcare, government, and education.
My Speech at the Conference
Towards the end of the one-hour-long event which I arrived late to attending only the last few minutes, I seized the opportunity to speak. Here is a recording and transcript of my speech:
"Hi, thank you. My name is Kevin Bihan-Poudec. I am the founder of a nonprofit organization called Voice For Change, which aims at advocating for ethical AI and job preservation. I am a displaced tech worker who has lost his job and been unable to reenter the tech sector in the last six months due to tech companies streamlining their operations to AI.
To stay on the topic of the day, I don't know if this was mentioned because I arrived late, but in the context of discussing discrimination, I would like to address immigration services. I'm originally from France, and I have been a victim of discrimination while trying to become an American citizen due to my sexuality.
Regarding the utilization of AI, I would like to open the question of how we can leverage this technology to prevent discrimination? AI could potentially look at the data impartially to determine qualifications and work in parallel with human decision-making to balance the process. My question is, how do you think we can leverage this technology to help with that?"
Using ChatGPT, I confirmed that I was technically eligible for American citizenship after establishing 4 years and 9 months of residency on a 10-year marriage green card.
Kevin:
How early can you apply for U.S. citizenship on a 10-year marriage green card? I believe you need to establish 5 years of residency to be eligible to apply, but due to the lengthy application process, is it true that you can apply after 4 years and 6 months of residency?
ChatGPT:
Yes, you are correct. If you have a 10-year marriage-based green card, you generally need to establish 5 years of continuous residency before you can apply for U.S. citizenship through naturalization. However, the U.S. Citizenship and Immigration Services (USCIS) allows you to apply up to 90 days before you meet the 5-year residency requirement.
This means you can submit your application after 4 years and 9 months of residency. It's important to ensure that you meet all other eligibility criteria, such as physical presence and good moral character, before applying.
After the USCIS immigration officer reviewed my application, which was in good standing due to my established lawful residency in the United States, they noted my history. I first came to the U.S. as a 16-year-old exchange student from France on a J-1 visa and later continued my studies for several years on an F-1 visa. In 2013, when same-sex marriage became legal again in California, I married my partner of five years. Despite this, when the officer saw my wedding photos and realized my marriage was same-sex, I was told I was ineligible for citizenship because I had applied "too early." Consequently, I was instructed to submit a written withdrawal of my citizenship application, as depicted in the reenactment above.
Advocating for Equity: The Role of AI in Eliminating Discrimination in Immigration
Extract of my Open Letter “Act Now” aimed at The U.S. Congress (page 73):
Section Title: Advocating for Equity: The Role of Al in Eliminating Discrimination in Immigration
In my own journey towards becoming an American citizen, I faced not just procedural hurdles but also discrimination that deeply impacted me. Despite fulfilling the prerequisites for application, I was unjustly told that I had applied too early and was denied citizenship —a pretext that obscured the real issue: discrimination based on my sexual orientation. This experience highlights the critical need for transparency, fairness, and objectivity in our immigration process—a goal that can be significantly advanced through the integration of artificial intelligence (AI). By leveraging Al to assess applications, we can minimize subjective biases and ensure that applicants are evaluated based on clear, equitable criteria.
Canada's foray into using machine learning for immigration decisions illustrates the potential of Al to expedite application processing while curtailing human biases. This approach aims for a system characterized by openness, friendliness, transparency, and inclusiveness. Further innovation is exemplified by the development of immigrant-designed tools like the chatbot Destin, which guides applicants through the immigration process, showcasing Al's ability to provide personalized, efficient assistance.
In the broader legal domain, Al is transforming immigration law by enabling the swift and accurate analysis of vast volumes of applications and legal documents. This technology can identify inconsistencies, missing information, and potential issues, significantly reducing the time attorneys spend on routine tasks and allowing them to concentrate on more complex case aspects. However, as we embrace Al's capabilities, it's imperative to address potential biases within these systems to ensure they offer fair and impartial treatment to all applicants.
Moreover, the use of artificial intelligence in immigration processes promises more than just efficiency; it is a vital step towards eliminating human error and the subjective biases that can unfairly influence outcomes. With Al's capability for impartial, data-driven analysis, we can create a system where decisions are made based on objective qualifications rather than personal characteristics. This approach not only makes the process more just but also upholds the values of equality and nondiscrimination at the heart of our immigration system.
Furthermore, employing artificial intelligence offers a powerful tool for reducing discrimination based on gender, sexual orientation, religion, origins, or color of skin. By designing algorithms that focus exclusively on the relevant criteria for citizenship, we can ensure a level playing field for all applicants, free from prejudice. This method aligns with our nation's principles of fairness and inclusivity, significantly reducing the likelihood of unjust denials. The adoption of Al in evaluating citizenship applications thus serves not only to streamline the process but also to reinforce our dedication to a […]
Extract of my Open Letter “Act Now” aimed at The U.S. Congress (page 74):
[…] more inclusive, respectful approach to immigration, where everyone is judged fairly and with dignity.
Making Connections and Building Bridges
After the conference, I had the opportunity to connect with a participant named Miguel, a student from the UCSD school, who approached me to learn more about my non-profit organization, Voice For Change. We exchanged information, and I'm excited to continue our conversation about leveraging AI for advocacy and to hear more about his studies and research.
Subject: Great meeting you at the AI Discrimination Conference in Downtown LA
Content of the email:
Hi Miguel,
It was a pleasure meeting you after the AI discrimination conference. I appreciated our brief conversation and your interest in my non-profit organization and advocacy efforts.
As I mentioned, my work focuses on ethical AI and job preservation, especially in the context of its impact on marginalized communities which is highlighted in my open letter “Act Now” aimed at the U.S Congress. Sharing my personal story about being denied citizenship due to my sexual orientation has been a driving force behind my efforts to highlight the broader societal implications of AI and automation.
I'd love to hear more about your studies and how you're leveraging AI in your work. It sounds like you're doing some fascinating research, and I'd be interested to know more about your focus and any insights you've gathered.
Let's stay in touch and explore how we can support each other's initiatives. Looking forward to hearing from you soon.
Best regards,
Kevin Bihan-Poudec
Founder Voice For Change Foundation | Advocating for Ethical AI and Job Preservation
info@voiceforchangefoundation.org
Conclusion
To ensure AI fosters equity in immigration processes, emphasizing ethical guidelines and regulatory frameworks is crucial. This approach aims to minimize subjective biases, ensuring decisions are based on clear, equitable criteria. By integrating AI, we can enhance transparency, fairness, and efficiency, making the immigration system more inclusive and respectful. Leveraging AI for impartial analysis promises not just streamlined processes but a commitment to fairness and nondiscrimination, ensuring all applicants are judged equally, aligning with the principles of fairness and inclusivity at the nation's core.
By sharing my personal experiences and advocating for the responsible use of AI, I hope to contribute to a more equitable future where technology serves as a tool for justice and inclusivity.
