1. Definitions:
- AI System: Any system that is designed by humans with the intention to perform tasks that if done by a human would require intelligence. Such systems include those capable of learning, adapting, and improving their performance based on experience or data.
- High-Risk AI System: Any AI system that can cause significant harm to individuals or society either through direct interaction or decision-making that can affect individuals or society at large.
- Data Privacy: The right of individuals to control or influence what information related to them may be collected and stored, and by whom and to whom that information may be disclosed.
2. Transparency and Accountability:
- All AI systems must have clear and accessible information explaining their purpose, capabilities, and limitations.
- High-Risk AI systems must maintain a detailed log of their operations to facilitate audits.
- Organisations deploying AI must have a responsible party for every deployed AI system.
3. Privacy and Data Protection:
- AI systems must comply with the existing data protection laws, ensuring user data is not used without informed consent.
- Data used by AI systems must be anonymised to protect user identity.
- AI systems must provide users with an option to opt-out of data collection procedures.
4. Fairness and Non-discrimination:
- AI systems must not foster or perpetuate any form of discrimination or bias.
- All AI system development and deployment must take measures to minimise bias, ensuring equitable outcomes for all users.
5. Safety and Security:
- All AI systems should meet defined safety standards.
- AI systems must have robust, reliable security measures in place to prevent misuse and protect against hacking.
- High-risk AI systems must undergo stringent safety and security checks before deployment.
6. Human Oversight:
- AI systems should not undermine human autonomy or cause harm.
- There should be a level of human oversight on high-risk AI systems, including an option for manual override.
7. Intellectual Property Rights:
- Clear guidelines must be developed regarding ownership and rights over AI-created content.
- The use of copyrighted data for training AI systems must be regulated.
8. Education and Public Awareness:
- The government should invest in educating the public about AI, its uses, and implications.
- Public awareness campaigns should be undertaken to ensure that the society understands AI and its impact.
9. Environmental Impact:
- The environmental impact of AI systems, such as energy consumption and electronic waste, should be considered and minimised.
10. Regulatory Authority:
- A regulatory authority will be established to enforce this act, regulate AI activities, conduct audits, and ensure compliance.
- The regulatory body will have the power to impose sanctions for non-compliance, including fines and operational bans.
In Conclusion
And ChatGPT concludes, “This act seeks to strike a balance between fostering technological innovation and ensuring the protection and welfare of individuals and society at large”
