Risk Identification and Mitigation in AI
- Morgan Hunter
- Feb 13
- 5 min read
Into the Shadows with Erica Simmons
In this episode of Monster in My Closet, Anil and Morgan explore the risks of rogue AI, guided by Erica Simmons' expert insights. Together, they delve into real-world examples, such as biased AI in law enforcement, and discuss the critical need for ethical considerations, human oversight, and regulatory guardrails in AI development. They tackle questions about balancing innovation with societal responsibility, the potential for AI to cause job displacement, and how personalized AI systems might mitigate some of these challenges. The conversation underscores the importance of centering people in decision-making and leveraging AI for good while minimizing harm. These show notes contain:
"AI is like the dumpster of the internet—if you put trash in, you get trash out."
Key Takeaways
Ensure Oversight in AI Use: Always maintain human oversight when deploying AI systems, especially in high-risk use cases like law enforcement or healthcare. Assign a team for AI supervision. Regularly audit AI-generated decisions for accuracy. Establish feedback loops with developers for errors or anomalies.
Implement Ethical AI Practices: Create or enhance governance structures to oversee AI ethics. Hire or train individuals with ethical expertise, not just compliance knowledge. Develop an AI ethics charter for your organization. Include ethics reviews in procurement and development stages.
Ask Tough Questions of Vendors: Before adopting AI systems, scrutinize the data and algorithms. Request transparency on data sources and error margins. Use procurement checklists to evaluate AI providers. Avoid vendors unable to provide comprehensive answers.
Prepare Data for AI Integration: Ensure your organization's data is clean and structured. Invest in data cleaning and organization. Conduct periodic data integrity audits. Involve diverse teams, including neurodiverse talent, for detailed analysis.
Focus on People-Centric Decisions: Avoid prioritizing cost savings over societal impact. Use "what-if" scenarios to anticipate AI's impact on employees. Engage stakeholders in discussions about ethical trade-offs. Support displaced workers with retraining and upskilling programs.
Research and Statistics
AI in Law Enforcement: Studies show AI-driven facial recognition systems are up to 100 times more likely to misidentify people of color compared to white individuals.
AI Job Displacement Risks: The World Economic Forum predicts AI will replace 85 million jobs by 2025, but it will also create 97 million new roles, emphasizing the need for reskilling.
Regulatory Developments: The EU’s AI Act introduces legal frameworks to regulate AI and prevent harm, influencing global governance practices.
Ethics in AI Adoption: A McKinsey report highlights that only 25% of companies have policies governing AI ethics, exposing organizations to reputational and operational risks.
Suggested Reading
Weapons of Math Destruction by Cathy O’Neil
Algorithms of Oppression by Safiya Umoja Noble
AI Ethics by Mark Coeckelbergh
The Age of AI by Henry A. Kissinger, Eric Schmidt, and Daniel Huttenlocher
Revenge of the Tipping Point by Malcolm Gladwell
These Women Warned Of AI's Dangers And Risks Long Before ChatGPT: https://www.rollingstone.com/culture/culture-features/women-warnings-ai-danger-risk-before-chatgpt-1234804367/
Wrap-up
AI holds immense promise, but its unchecked application poses significant risks to society. By prioritizing human oversight, demanding transparency from AI vendors, and implementing robust ethical frameworks, organizations can harness AI's potential while minimizing harm. The challenge lies in balancing innovation with responsibility, ensuring that AI serves humanity without compromising fairness, safety, or livelihoods. As Erica Simmons reminds us, putting people at the center of decisions is not just a corporate imperative—it’s a moral one.
About Erica Simmons
Erica Simmons is a dynamic innovator with a passion for driving transformative experiences in the tech space. she is an expert in product development, business development and marketing enterprise software having gained extensive experience from her time at some of the top software companies in the world notably ServiceNow, Siemens AG and Oracle Corporation. Her expertise in emerging technologies has guided businesses, governments, and educational institutions to adapt to the rapidly changing landscape of work.
A key area of her influence has been in advocating for policies that ensure the Small Island Developing States like those in the Caribbean and around the world are well-prepared for the future of work. Ms. Simmons has tirelessly promoted initiatives that integrate AI into the redevelopment strategies, ensuring equitable access to technology and emphasizing the need to address infrastructure, cultural, and governmental hurdles.
As one of the global voices on AI, she serves on UNESCO’s Expert Group on the Ethics of AI and is a part of UNESCO’s AI Ethics Experts Without Borders Network, contributing to frameworks that guide the responsible adoption of AI around the world. Most notably, she’s the author of the Caribbean Artificial Intelligence Roadmap, a strategic framework aimed at fostering innovation, building infrastructure, and ensuring the ethical use of AI across the Caribbean.
A highly sought-after public speaker, Erica shares her expertise on the future of work, AI, and the evolving world of technology. Get ready for an insightful conversation with one of the industry’s leading voices on ethical AI and digital transformation.
Resources from Erica:
UNESCO website: Caribbean Artificial Intelligence Policy Roadmap.
UNESCO's first-ever global standard on AI ethics – the ‘Recommendation on the Ethics of Artificial Intelligence’, adopted in 2021, is applicable to all 194 member states of UNESCO.
The Information for All Programme (IFAP) is a unique UNESCO intergovernmental programme that supports Member States in fostering equitable societies through promoting universal access to information and knowledge for sustainable development.
A platform for knowledge, expert insights, and good practices on the ethics and governance of artificial intelligence
The Caribbean AI Initiative, addresses the critical need for shaping an ethical and inclusive AI landscape in the Caribbean SIDS.
The overall purpose of the Caribbean Artificial Intelligence (AI) initiative is to raise awareness and promote reflection on the opportunities and challenges that Artificial Intelligence and its correlated technologies pose in the Caribbean Small Island Developing States (SIDS).
Ethics, Equality, & AI in the Caribbean
Artificial intelligence (AI) will affect the socio-economic development of nations across the globe. Caribbean countries are particularly susceptible because they tend to be labor intensive economies and are therefore at risk of significant economic and social disruption from automation and artificial intelligence. Three experts in this space--Cordel Green, Stacey Russell, and Erica Simmons--discuss these issues and much more.
AI Risk Management Framework
NIST released NIST-AI-600-1, Artificial Intelligence Risk Management Framework: Generative Artificial Intelligence Profile. Developed in part to fulfill an October 30, 2023 Executive Order, the profile can help organizations identify unique risks posed by generative AI and proposes actions for generative AI risk management that best aligns with their goals and priorities.
ISO/IEC 42001:2023
Information technology — Artificial intelligence — Management system
Ethically Aligned Design
ETHICALLY ALIGNED DESIGN – IEEEA Vision for Prioritizing Human Well-being with Autonomous and Intelligent Systems
Comments