Conclusion
This page aims to wrap up the website and remind you of some key takeaways. It further acts to offer potential solutions to the data privacy crisis we are heading towards.
Concerns
- Companies use various methods of data mining, such as web scraping and eavesdropping, to train their models.
- The data used often includes a degree of your personal information and the methods used to obtain it often violate your privacy rights.
- Meanwhile, information on your privacy rights can be taxing to find and it can be even more difficult to determine when they are violated.
Cookies and Privacy Policies
- First party cookies are website specific; whereas, third party cookies track your data across multiple websites.
- You sign away your data via accepting all cookies on websites and accepting terms and conditions on apps.
- The consolidation of your data via AI models makes it easier to steal.
User Data Impact on AI
- Companies use your data to train AI models.
- Models created through personal data can contain biases harmful to your community.
- Similarly, companies engage in specialized marketing with your data, and job reduction through chat-bots taught on human interaction.
Source: Rightly
Ways to Protect Your Data
- Keep Confidential Information Offline
- Read Privacy Policies!!
- Learn and Enact your Data Rights
Federated Learning- As put by Ferm and co., Federated Learning is "a privacy-focused approach to machine learning where algorithms collaborate without sharing user data itself.” Ideally, the decentralization of data eliminates some of the risk associated with data leaks.
Responsible AI Standard V2- Jayachandran mentions RAISV2, a comprehensive guide for building AI systems responsibly. It aims to guide product development toward more beneficial and equitable outcomes. By focusing on fairness, reliability, privacy, security, inclusiveness, transparency, and accountability, AI companies become significantly less invasive.
Open Responsible AI Licensing- Jayachandran also introduces a current trend in AI ethics, which aims to incorporate the openness of traditional software licenses to ensure AI is used ethically. A noble idea, but whether it has much impact can only be determined by time.
Data governance- According to Jernite et al., Data governance is the management of data availability, usability, integrity, and security in a system. It emphasizes informed consent, data subject rights, and responsible data use. This solution requires companies to request explicit permission from individuals whose data is collected without their knowledge, ensuring said data is used only for intended purposes while respecting privacy and confidentiality.
Responsible AI Standard V2- Jayachandran mentions RAISV2, a comprehensive guide for building AI systems responsibly. It aims to guide product development toward more beneficial and equitable outcomes. By focusing on fairness, reliability, privacy, security, inclusiveness, transparency, and accountability, AI companies become significantly less invasive.
Open Responsible AI Licensing- Jayachandran also introduces a current trend in AI ethics, which aims to incorporate the openness of traditional software licenses to ensure AI is used ethically. A noble idea, but whether it has much impact can only be determined by time.
Data governance- According to Jernite et al., Data governance is the management of data availability, usability, integrity, and security in a system. It emphasizes informed consent, data subject rights, and responsible data use. This solution requires companies to request explicit permission from individuals whose data is collected without their knowledge, ensuring said data is used only for intended purposes while respecting privacy and confidentiality.
Potential Solutions to AI Invasiveness
