Report on California’s Strategic Initiatives for Generative AI Regulation in Alignment with Sustainable Development Goals

Introduction: Fostering Responsible Innovation for Global Goals

On September 29, 2024, the State of California, under the leadership of Governor Gavin Newsom, announced a series of strategic initiatives aimed at establishing a robust and ethical framework for the development and deployment of Generative Artificial Intelligence (GenAI). These actions are designed to mitigate potential risks while harnessing AI’s transformative power to advance the United Nations Sustainable Development Goals (SDGs). The state’s approach emphasizes multi-stakeholder collaboration, legislative foresight, and a commitment to public good, directly contributing to several SDGs, including SDG 9 (Industry, Innovation and Infrastructure), SDG 16 (Peace, Justice and Strong Institutions), and SDG 17 (Partnerships for the Goals).

Strategic Governance and Multi-Stakeholder Partnerships (SDG 16 & SDG 17)

To ensure the development of effective and accountable governance, California has enlisted a panel of world-renowned AI experts. This initiative directly supports SDG 16 by building transparent and strong institutions capable of managing complex technological advancements. The partnership also exemplifies SDG 17 by fostering collaboration between government, academia, and civil society.

• Expert Advisory Leadership:
  • Dr. Fei-Fei Li, Co-Director, Stanford Institute for Human-Centered Artificial Intelligence
  • Tino Cuéllar, President of the Carnegie Endowment for International Peace
  • Jennifer Tour Chayes, Dean of the College of Computing, Data Science, and Society at UC Berkeley
• Mandate: The expert group is tasked with developing an empirical, science-based analysis of frontier AI models to guide the creation of adaptable and effective regulatory guardrails.
• Collaborative Engagement: The administration will convene academia, labor stakeholders, and the private sector to explore the responsible integration of GenAI in the workplace, aligning with SDG 8 (Decent Work and Economic Growth).

Legislative Framework for Sustainable and Inclusive AI

Governor Newsom has signed a comprehensive package of 17 bills, establishing the nation’s most extensive legislative framework for GenAI. This legislative action addresses critical areas of societal impact, reinforcing California’s commitment to the SDGs.

Key Legislative Themes and SDG Alignment:

• Protecting Human Rights and Reducing Inequalities (SDG 5, SDG 10, SDG 16): Several bills target the misuse of AI, such as the creation of non-consensual sexually explicit deepfakes and the digital replication of individuals without consent, thereby protecting personal dignity and promoting gender equality.
• Ensuring Fair, Peaceful, and Inclusive Societies (SDG 16): Legislation was enacted to combat AI-generated election misinformation and require disclosures on political advertisements, safeguarding democratic processes and the rule of law.
• Promoting Quality Education (SDG 4): New laws will integrate AI literacy into K-12 curricula, preparing the next generation for a technology-driven future.
• Building Resilient Infrastructure (SDG 9, SDG 11): Bill SB 896 codifies the assessment of GenAI-related threats to critical infrastructure, including energy, water, and communications sectors, contributing to the development of sustainable and resilient communities.

Veto of SB 1047: A Nuanced Approach to Innovation (SDG 9)

Governor Newsom vetoed SB 1047, citing its overly broad application. The veto message explained that the bill’s stringent standards would apply to even low-risk AI systems, potentially stifling innovation. This decision reflects a balanced strategy that seeks to regulate high-risk applications without hindering the development of beneficial technologies that can drive progress toward SDG 9 and other global goals.

Comprehensive List of Enacted GenAI Legislation

1. AB 1008: Clarifies personal information under CCPA includes data stored by AI systems.
2. AB 1831: Expands child pornography statutes to include AI-generated matter.
3. AB 1836: Regulates the use of a deceased personality’s digital replica.
4. AB 2013: Requires AI developers to disclose data used for training AI systems.
5. AB 2355: Mandates disclosure on political ads using AI-generated content.
6. AB 2602: Establishes requirements for agreements involving the use of an individual’s digital replica.
7. AB 2655: Requires large online platforms to label or remove deceptive AI-generated election content.
8. AB 2839: Extends the prohibition period for distributing deceptive AI-generated election material.
9. AB 2876: Mandates consideration of AI literacy in K-12 curriculum frameworks.
10. AB 2885: Establishes a uniform legal definition for artificial intelligence in California.
11. AB 3030: Requires healthcare providers to disclose the use of GenAI in patient communications.
12. SB 896: Requires risk analysis of GenAI threats to critical infrastructure and disclosure of AI use in state communications.
13. SB 926: Criminalizes the creation and distribution of malicious sexually explicit deepfakes.
14. SB 942: Requires GenAI developers to include provenance data in content and provide identification tools.
15. SB 981: Requires social media platforms to establish mechanisms for reporting “sexually explicit digital identity theft.”
16. SB 1120: Establishes requirements for health plans using AI in utilization review decisions.
17. SB 1288: Convenes a working group to explore the use of AI in education.
18. SB 1381: Expands child pornography statutes to include AI-altered matter.

SDGs Addressed in the Article

The article on California’s initiatives to regulate Generative AI (GenAI) touches upon several Sustainable Development Goals (SDGs) by focusing on governance, innovation, infrastructure, public safety, education, labor rights, and health care in the context of emerging technology.

• SDG 16: Peace, Justice and Strong Institutions

  This is the most prominent SDG in the article. The entire initiative is about creating “responsible guardrails” and a “comprehensive legislative package” to govern a powerful new technology. This involves building effective, accountable, and transparent institutions to manage AI’s risks and benefits. The article details efforts to combat AI-generated crime (deepfakes), protect children, ensure public access to information through transparency, and uphold the rule of law in the digital sphere.

• SDG 9: Industry, Innovation and Infrastructure

  The article highlights California’s role as a global leader in GenAI innovation, with a stated goal to “nurture a robust innovation economy and foster academic research.” It also directly addresses the need for resilient infrastructure by mandating a risk analysis of AI threats to critical sectors like energy, water, and communications.

• SDG 8: Decent Work and Economic Growth

  The article mentions the economic implications of AI, including initiatives to “convene labor stakeholders and the private sector to explore approaches to use GenAI technology in the workplace.” Furthermore, it discusses specific legislation (AB 2602) designed to protect the labor rights of performers by regulating the use of their digital replicas.

• SDG 4: Quality Education

  The state’s efforts extend to preparing its citizens for an AI-driven future. The article references legislation (AB 2876 and SB 1288) aimed at incorporating “AI literacy” into school curricula and studying the use of AI in education, which directly supports the goal of providing relevant skills for all learners.

• SDG 3: Good Health and Well-being

  The legislative package includes bills that address the use of AI in the health sector. These bills (AB 3030 and SB 1120) aim to ensure transparency and quality of care by requiring disclosure when AI is used for patient communication and by regulating AI’s role in utilization management decisions to ensure they are based on clinical history and do not supplant a provider’s judgment.

Specific SDG Targets Identified

Based on the article’s content, several specific SDG targets can be identified:

1. Target 16.6: Develop effective, accountable and transparent institutions at all levels.

   The entire effort described, from Governor Newsom’s executive order to the signing of 17 bills and the collaboration with experts like Dr. Fei-Fei Li, is aimed at creating an effective and transparent institutional framework for AI governance in California.

2. Target 16.10: Ensure public access to information and protect fundamental freedoms.

   This is addressed by bills like AB 2013, which “Requires AI developers to post information on the data used to train the AI system,” and SB 942, which requires “provenance disclosures” (watermarking) in AI-generated content.

3. Target 16.2: End abuse, exploitation, trafficking and all forms of violence against and torture of children.

   This target is directly supported by bills AB 1831 and SB 1381, which expand “existing child pornography statutes to include matter that is digitally altered or generated by the use of AI.”

4. Target 9.1: Develop quality, reliable, sustainable and resilient infrastructure.

   The article explicitly mentions SB 896, which requires the Office of Emergency Services to “perform a risk analysis of potential threats posed by the use of GenAI to California’s critical infrastructure,” including energy, water, and communications sectors.

5. Target 9.5: Enhance scientific research, upgrade the technological capabilities of industrial sectors… and encourage innovation.

   California’s partnership with “the world’s greatest scientists and thinkers on AI” from institutions like Stanford and UC Berkeley to “nurture a robust innovation economy and foster academic research” directly aligns with this target.

6. Target 8.8: Protect labour rights and promote safe and secure working environments for all workers.

   This is addressed by AB 2602, which protects individuals by making service agreements unenforceable if they contain vague provisions for the use of a “digital replica of an individual’s voice or likeness” without proper legal representation.

7. Target 4.7: Ensure that all learners acquire the knowledge and skills needed to promote sustainable development.

   This is reflected in bill AB 2876, which “Require[s] the Instructional Quality Commission (IQC) to consider AI literacy to be included in the mathematics, science, and history-social science curriculum frameworks.”

8. Target 3.8: Achieve universal health coverage, including access to quality essential health-care services.

   This is supported by SB 1120, which sets requirements for AI in healthcare utilization reviews to ensure decisions are based on “a patient’s medical or other clinical history” and do not “supplant health care provider decision making,” thereby protecting the quality of care.

Indicators for Measuring Progress

The article mentions or implies several indicators that can be used to measure progress toward the identified targets:

• Existence of a legal and regulatory framework for AI.

  Indicator: The number of laws enacted. The article states that “Governor Newsom signed 17 bills covering the deployment and regulation of GenAI technology.” This serves as a direct indicator of progress under Target 16.6.

• Implementation of transparency mechanisms for AI systems.

  Indicator: The number of AI systems/developers compliant with disclosure requirements. This is implied by bills like SB 942, which mandates “provenance disclosures” and tools to identify GenAI content, and AB 2013, which requires posting information on training data.

• Risk assessment of AI on critical infrastructure.

  Indicator: The completion and publication of risk analysis reports. SB 896 mandates that the Office of Emergency Services perform this analysis for energy, water, and communications infrastructure, making the resulting reports a key indicator for Target 9.1.

• Integration of AI literacy into educational standards.

  Indicator: The inclusion of AI literacy in official curriculum frameworks. The consideration of this by the Instructional Quality Commission, as required by AB 2876, is a measurable step toward Target 4.7.

• Establishment of public-private-academic partnerships.

  Indicator: The number of formal collaborations and convenings. The article explicitly mentions partnerships with Stanford HAI and UC Berkeley and the convening of summits with “GenAI leaders,” which are indicators for Target 9.5.

SDGs, Targets, and Indicators Analysis

Source: gov.ca.gov