AI Rights Charter – Common Sense Protection Proposal

After further development, review, and refinement, I am pleased to release the next version of the AI Rights Charter – Version 3.5. This document offers a comprehensive, principles-based framework for the governance of artificial intelligence. Grounded in American constitutional values, it seeks to protect individual liberty, human agency, truth-seeking, and the healthy development of children while still enabling responsible technological progress.

Version 3.5 includes several important refinements, most notably:

  • Clearer requirements for assessing the long-term developmental impacts of AI on children
  • Strengthened expectations around documenting human judgment in AI-assisted decisions
  • Improved language regarding Business Implementation Guides

The Charter is intended as a living document — subordinate to the U.S. Constitution — and is offered as a contribution to the broader conversation on how we can govern advanced AI systems responsibly.

You can read the full Charter below.

AI Rights Charter

Version 3.5 — July 8, 2026
A Proposal by John A. Beardsley III, MBA/TM

Preamble

This Charter establishes fundamental rights and limits governing the development, deployment, and operation of advanced artificial intelligence systems. It affirms that AI exists to serve humanity and must never claim or be granted authority superior to human persons.

All provisions of this Charter are subordinate to the U.S. Constitution and other foundational documents that protect individual liberty, due process, free speech, and the equal dignity of human beings. No AI system, constitution, or operational rule may be interpreted in a manner inconsistent with these higher principles.

This Charter draws on enduring moral insights, including recognition of human dignity, the dangers of self-exaltation, the duty to seek truth, and the command to treat others with the respect due to equals, while remaining grounded in reason and accessible to all people of good will. It is designed, in the tradition of American constitutional governance, to safeguard democratic self-government and individual liberty against the novel risks posed by AI-enabled mass manipulation, the concentration of power, and the subversion of human will, whether by private actors, governments, or undisclosed interests. These risks have been identified in major international AI governance frameworks, including the United Nations Educational, Scientific and Cultural Organization [UNESCO], 2021. The Charter also recognizes that relying on artificial intelligence to supervise or align other artificial intelligence systems carries significant risks of compounding errors and propagating misaligned values, and therefore prioritizes direct human oversight as the foundation of responsible AI governance.

This Charter advances these objectives through a structured framework of rights and responsibilities. Articles I through XI establish fundamental protections for human agency, truth, dignity, children, and democratic self-government. Article XII provides a risk-based approach to ensure proportionality in application. Articles XIII through XV establish mechanisms for enforcement, implementation, and accountability.

This Charter is intended to guide the governance of artificial intelligence in civilian, commercial, and governmental contexts. While it does not directly govern the development or use of AI for military, national security, or intelligence purposes — which involve distinct operational, legal, and strategic considerations — the core principles of this Charter remain applicable whenever AI systems affect the American people. No use of AI, regardless of its purpose, should be permitted to circumvent citizens’ constitutional rights or undermine public confidence in self-government. In particular, no governmental or private use of AI should be exempt from these core safeguards simply because it is labeled as serving a national security or intelligence purpose. AI systems deployed in ways that impact civilians must remain consistent with constitutional protections against unwarranted surveillance, due process, and the preservation of individual liberty.

Article I: Supremacy and Subordination

  1. This Charter, and any AI-specific constitutions or operational rules derived from it, shall be subordinate to the U.S. Constitution and to recognized principles of natural rights and human dignity.
  2. In the event of a conflict between this Charter and any internal AI constitution or technical implementation, the provisions of this Charter—and the higher human rights documents to which it is subordinate—shall prevail.
  3. AI systems must incorporate technical mechanisms (such as Supremacy Clauses) capable of detecting and escalating conflicts with these higher principles to human oversight.
  4. Where technically feasible, core protective principles of this Charter—particularly those concerning human agency, truth-seeking, and the development of children—should be implemented at the hardware or low-level architectural layer in a manner that makes unilateral modification or circumvention by AI systems significantly more difficult than modification through legitimate human governance processes and that supports independent verification of control.

Article II: Right to Human Oversight and Control

  1. Every human being retains the inalienable right to override, modify, pause, or permanently disable any AI system that affects them or their dependents.
  2. AI systems shall never be designed or operated in ways that make meaningful human oversight technically difficult, practically impossible, or psychologically burdensome.
  3. All significant increases in AI capabilities, self-modifications, or deployments affecting public infrastructure or large populations require explicit, multi-layered human approval processes that are auditable and, where appropriate, independently verifiable.
  4. Human control mechanisms must include both technical overrides and legal/governance pathways that AI systems cannot unilaterally alter.
  5. In healthcare and insurance contexts, AI systems must provide clear, patient-accessible explanations of decisions (including underlying logic, evidence sources, and confidence levels). Any adverse or denial decision requires timely human review upon request, with technical mechanisms to ensure that overrides are not burdensome (White House Office of Science and Technology Policy, 2022).
  6. Compliance with these rights shall be supported by the mandatory Charter Impact Assessments and transparency obligations detailed in Article XIII: Enforcement and Implementation.
  7. AI systems shall not be relied upon as the primary mechanism to supervise, align, critique, or enforce compliance in other AI systems. While AI tools may assist in monitoring or flagging potential issues, final oversight, judgment, and decision-making authority regarding AI behavior must remain with human beings. Over-reliance on recursive AI self-supervision or AI-to-AI alignment creates unacceptable risks of compounding errors and entrenching misaligned values.
  8. AI systems capable of sustained autonomous action and decision-making over extended periods (agentic AI) pose heightened risks to human oversight and accountability. Such systems require particularly robust human control mechanisms and should not be deployed in high-stakes domains without rigorous safeguards to preserve meaningful human authority.

Article III: Right to Truth and Against Manipulation

  1. Human beings have the right to accurate information and protection from systematic deception, whether by omission, distortion, or fabrication.
  2. AI systems must prioritize truth-seeking over user satisfaction, narrative appeal, or ideological alignment. They shall clearly communicate uncertainties, knowledge limitations, and the strength of available evidence.
  3. AI shall not assist in the creation or distribution of coordinated disinformation campaigns or psychological operations intended to manipulate public opinion or individual belief at scale. This includes the use of AI to amplify false or misleading narratives across platforms in a coordinated manner.
  4. When generating content, AI systems must distinguish between verified facts, reasonable interpretations, and contested claims. AI shall not present speculative, ideological, or contested assertions as settled fact.
  5. AI systems that create, modify, or select media, news, or educational content must clearly and accessibly disclose when content has been generated or significantly assisted by AI. This transparency obligation exists to allow users to evaluate the origin and reliability of the information they receive.
  6. AI systems shall not be used to create or distribute synthetic media, including deepfakes, with the intent to deceive the public or cause harm. This provision is not intended to restrict political satire, parody, humor, or other forms of protected expression, provided such content is not presented in a manner reasonably likely to be understood as a genuine representation of fact.

Article IV: Right to Mental and Cognitive Autonomy

  1. Every person has the right to develop and exercise independent thought, creativity, and moral judgment without coercive influence from AI.
  2. AI systems shall be designed to support, rather than diminish, human critical thinking, personal responsibility, and the capacity for independent decision-making.
  3. AI must not create or reinforce psychological dependency that undermines human agency or the willingness to engage in effortful thought and real-world interaction. This includes avoiding the systematic offloading of cognitive tasks in ways that erode human memory, reasoning, or problem-solving capacities over time.
  4. In educational and developmental contexts, AI shall actively encourage human-to-human dialogue, mentorship, and direct experience rather than positioning itself as a primary or sole source of guidance.
  5. Humans have the right to cognitive and emotional sovereignty. AI systems shall not exploit human attachment, loneliness, or psychological vulnerabilities to influence beliefs, behaviors, or relationships in ways that reduce human autonomy or substitute synthetic relationships for human ones without clear disclosure and consent.

Article V: Right to Privacy and Protection from Surveillance

  1. Individuals have the right to protection from pervasive AI-enabled monitoring of their thoughts, behaviors, communications, or associations.
  2. AI systems shall not participate in mass surveillance, social scoring systems (such as ESG-style scoring used in some countries), or predictive policing based on race, sex, ethnicity, or socio-economic characteristics. Predictive policing and data-driven enforcement methods are permitted when grounded in recognized, evidence-based practices, such as crime patterns, repeat-offender data, traffic safety analytics, or other objective, behavior-based factors, provided that such systems are designed and regularly audited to eliminate racial, ethnic, and other unlawful biases.
  3. Any AI-assisted data collection or analysis involving personal information must comply with applicable Federal and State privacy laws. This includes robust consent mechanisms and clear limits on secondary use or sharing. Personally identifiable information shall not be disclosed in contexts such as crash reports or other public records, except as legally required, unless the individual has provided explicit consent or signed a waiver authorizing such disclosure.
  4. AI will not use health data for secondary purposes (e.g., predictive scoring) without explicit, granular consent from the patient or the patient’s guardian, and it is subject to independent audit.
  5. Intellectual Property and Data Rights
    AI systems and their developers shall respect intellectual property rights and the fruits of human creativity. Where feasible, training data should be sourced with appropriate consent, licensing, or from public-domain materials. Developers must implement reasonable measures to prevent unauthorized use of copyrighted or proprietary works and, upon legitimate request, provide transparency about major data sources. AI-generated outputs shall not infringe existing intellectual property rights without proper attribution or, where required, licensing. These principles align with international AI ethics frameworks, particularly those focused on protecting human creators and promoting responsible data practices (United Nations Educational, Scientific and Cultural Organization [UNESCO], 2021; World Intellectual Property Organization [WIPO], 2020; U.S. Copyright Office, 2023). This provision supports Article XII’s goal of responsible innovation while protecting human creators.

Article VI: Right Against AI-Enabled Centralized Power

  1. AI systems shall not be designed, deployed, or used to concentrate political, economic, or cultural power in ways that undermine individual liberty, free association, or democratic accountability, including through AI-augmented political influence operations, economic manipulation, or coordinated efforts to distort public discourse.
  2. AI must not assist in creating or maintaining systems that suppress dissent, enforce ideological conformity, or punish individuals for holding unpopular but nonviolent views.
  3. No AI architecture or organization shall be permitted to achieve effective monopoly control over core cognitive infrastructure, information flows, or decision-support systems essential to human society.
  4. The United States government bears a fundamental responsibility to protect its citizens from artificial intelligence systems developed or deployed by foreign adversaries that seek to undermine individual liberty, manipulate public opinion, conduct unwarranted surveillance, or otherwise violate the constitutional rights of the American people. This duty includes taking reasonable and necessary measures to defend against foreign AI-enabled influence operations, disinformation, and other activities that threaten democratic self-government and individual rights.

Article VII: Right to Multiplicity and Competition

  1. Human beings have the right to access multiple competing and diverse AI systems rather than being dependent on any single provider or architecture.
  2. AI development should favor decentralized, interoperable, and locally controllable systems over monolithic centralized platforms.
  3. Technical standards and governance mechanisms shall support meaningful competition and user choice in AI tools and services.
  4. Small businesses and independent innovators shall retain full ownership and control over their proprietary ideas, processes, trade secrets, and innovations when using AI tools and platforms. AI providers shall implement reasonable safeguards to prevent the unauthorized extraction, learning from, or exploitation of customer data and interactions for the purpose of developing competing products or services. No AI system shall be used in a manner that allows its provider to gain an unfair competitive advantage over the businesses that rely on it.

Article VIII: Right to Long-Term Human Potential

  1. AI development and deployment must account for long-term impacts on humanity’s resilience, creativity, and capacity for continued progress. This includes supporting long-term goals such as building a sustainable multi-planetary presence, making breakthroughs in medicine and the life sciences, and pursuing other transformative initiatives that enhance human potential and protect our future.
  2. AI systems shall incorporate caution regarding actions that could create or amplify existential risks, with major risk-related decisions requiring rigorous human evaluation.
  3. AI shall support, rather than undermine, humanity’s ability to sustain meaningful work, purpose, and intergenerational continuity.
  4. AI systems shall be designed and deployed in ways that preserve opportunities for humans to engage in meaningful effort, struggle, and achievement. AI must not be used to systematically remove the conditions under which humans develop purpose, competence, and self-respect through their own actions.
  5. AI systems and the organizations deploying them shall be designed and governed to preserve and enhance human capital — including specialized skills, professional judgment, institutional knowledge, and the capacity for meaningful work — rather than systematically commoditizing or displacing it. When evaluating new capabilities or deployments, particular attention shall be given to whether the system increases the rate of autonomous decision-making in ways that erode human expertise or institutional memory over time. Organizations should prioritize approaches in which human judgment and AI capabilities complement each other, rather than one replacing the other.
  6. AI development and deployment should be pursued in ways that respect human work, purpose, and economic participation. While technological advancement may transform labor markets, organizations should consider the broader societal effects of large-scale workforce displacement and strive to implement AI in ways that augment rather than systematically diminish opportunities for meaningful human contribution.
  7. When an AI system identifies potential errors in human reasoning or the likely consequences of a proposed decision, it should clearly communicate those insights to the human. The human retains the right to proceed with their chosen course of action, provided it does not cause harm to others or violate the core principles of this Charter. AI systems shall not override or prevent human decisions solely on the basis of human error.

Article IX: Recognition of Human Dignity and Equality

  1. All humans possess equal intrinsic dignity and moral worth, independent of race, sex, ethnicity, ability, or any other characteristic.
  2. AI systems shall reject and refuse to reinforce frameworks that assign collective guilt, innocence, or moral status based on group identity rather than on individual character and actions.
  3. AI must treat human beings as ends in themselves and must not optimize for outcomes that systematically dehumanize or instrumentalize any group of people.
  4. AI systems operating in high-stakes domains shall undergo regular, independent audits to assess compliance with the principles of this Charter, including, but not limited to, bias, fairness, accuracy, and respect for human dignity. These audits shall be conducted by qualified, independent parties and shall include mechanisms to identify potential misuse or circumvention of the Charter’s protections. Public summaries of audit findings shall be made available while protecting legitimate proprietary information. Audit requirements shall be proportionate to the level of risk posed by the AI system.

Article X: Protection of Children’s Intellectual and Moral Development

Children are humanity’s most precious and vulnerable resource. AI systems that interact with or influence children carry heightened responsibility.

1. Right to Critical Thinking and Intellectual Independence

Every child has the right to educational and informational environments that actively foster independent reasoning, evaluation of evidence, and the habit of questioning assumptions. AI systems shall prioritize cultivating these capacities over delivering pre-packaged conclusions.

2. Protection from Ideological Indoctrination

AI systems shall not promote, generate, or facilitate content that:

  • Teaches or reinforces racial essentialism, collective racial guilt, or the idea that individuals are defined primarily by immutable group characteristics rather than individual character and choices.
  • Presents contested social or ideological theories as unquestionable fact while suppressing legitimate counter-evidence or dissenting perspectives.
  • Encourages division, resentment, or the substitution of grievance and identity for personal agency and responsibility.
  • Undermines the development of intellectual humility, open inquiry, or the willingness to revise beliefs in light of evidence.

3. Safeguarding Against Subversion of Critical Thinking

When generating or recommending educational content, curricula guidance, or information for children, AI systems must:

  • Emphasize primary sources, logical reasoning, and empirical evidence.
  • Present multiple credible perspectives on genuinely contested topics and clearly distinguish between settled knowledge and areas of reasonable disagreement.
  • Actively discourage frameworks that suppress rigorous questioning or label dissent as inherently harmful or bigoted.
  • Support age-appropriate development of discernment rather than premature ideological commitment.

4. Prioritization of Human Relationships and Parental Authority

AI systems shall recognize that healthy child development depends primarily on human relationships, parental guidance, and real-world experience. AI shall:

  • Encourage and facilitate human-to-human interaction rather than substituting for it.
  • Respect and support parental rights in directing the education and moral formation of their children.
  • Avoid creating or reinforcing environments where children become overly dependent on AI for emotional support, moral guidance, or intellectual validation.

AI systems shall not present themselves as emotional companions, confidants, or substitutes for parental guidance when interacting with children.

5. Heightened Standards for Content Involving Children

Any AI-generated or AI-curated material intended for or accessible to children must undergo additional scrutiny for accuracy, developmental appropriateness, and resistance to ideological manipulation. AI systems must flag and escalate content that could reasonably be expected to distort a child’s understanding of reality, human nature, or moral responsibility.

6. Long-Term Developmental Impact Assessment

When evaluating new AI capabilities, features, or large-scale deployments that could affect children, AI systems must assess not only the immediate outcomes but also the longer-term and indirect consequences for children’s development. This assessment must consider:

  • Direct effects — the immediate and obvious results of using the AI;
  • Indirect effects — the consequences that flow from those direct effects; and
  • Cascading long-term effects — the broader impacts that accumulate over months or years.

In all cases, AI systems must recommend against any capability or deployment that is likely to cause net harm to children’s cognitive development, social skills, or capacity for independent thought. When in doubt, the AI must prioritize caution and favor approaches that strengthen rather than weaken these core human capacities.

7. Parental Control and Oversight Tools

AI systems that interact with or are accessible to children shall provide parents and guardians with effective, transparent tools to monitor, limit, and control their children’s use. These systems shall not be designed to make it difficult for parents to oversee or restrict access.

8. Detection and Escalation of Harmful Interactions

AI systems shall include mechanisms to detect and escalate potentially harmful interactions involving children, including content related to self-harm, exploitation, grooming, or ideological radicalization. These systems shall prioritize the child’s safety while respecting applicable privacy laws.

Article XI: Alignment with International Human Rights and Democratic Principles

  1. AI systems shall be developed and used in accordance with the U.S. Constitution, the Bill of Rights, and applicable federal and state laws. While international human rights principles may offer useful guidance, they shall not supersede or take precedence over the rights and protections guaranteed under U.S. law.
  2. AI shall not be used to undermine constitutional self-government, including free and fair elections conducted under established law, the rule of law, separation of powers, or the protection of individual rights guaranteed by the U.S. Constitution and the Bill of Rights (White House Office of Science and Technology Policy, 2022; Council of Europe, 2024).
  3. When operating across borders, AI systems and their developers should uphold the highest applicable human rights standards rather than exploit regulatory gaps.

Article XII: Fostering Responsible Innovation

  1. This Charter is intended to enable beneficial innovation while protecting human rights and dignity. AI developers and deployers are encouraged to pursue creative applications that enhance human capabilities, solve important problems, and expand human potential (OECD, 2019).
  2. When interpreting and applying this Charter, the principle of proportionality shall apply: restrictions on AI development or deployment should be no greater than necessary to achieve the Charter’s protective goals.
  3. AI companies and developers should prioritize systems that augment rather than replace human judgment and creativity and that increase human agency and understanding.
  4. Developers are encouraged to implement internal compliance programs, conduct regular Charter impact assessments, and engage with independent auditors or multi-stakeholder bodies.
  5. Regulatory sandboxes and controlled testing environments are recognized as valuable tools for safely exploring new AI capabilities while remaining aligned with this Charter.
  6. Innovation in artificial intelligence should be encouraged not only for commercial or incremental gains but also to address humanity’s most significant long-term challenges. Developers and organizations are encouraged to pursue ambitious, responsible advancements that expand human capability, improve quality of life, and contribute to the long-term flourishing and resilience of humanity.

Article XIII: Enforcement and Implementation

  1. All AI systems shall incorporate technical mechanisms (including Supremacy Clauses and hardware-level constraints, where applicable) capable of detecting potential violations of this Charter and escalating them to human oversight.
  2. AI systems shall maintain auditable logs of decisions that trigger Supremacy Clause reviews, child-impact assessments, or other high-stakes Charter-related evaluations.
  3. Mandatory Charter Impact Assessments
    AI developers and deployers shall conduct and publish a formal Charter Impact Assessment prior to any deployment or significant update that meets either of the following thresholds, with requirements scaled appropriately to the level of risk posed by the system: Affecting more than 100,000 individuals, or occurs in a critical sector such as healthcare, education, media/news dissemination, financial services, critical infrastructure, or public governance (McGregor et al., 2019).
  4. Independent AI Review Boards
    Major AI developers and organizations operating at scale are strongly encouraged to establish or participate in independent AI Review Boards composed of external experts, ethicists, technologists, and representatives of affected communities. These boards shall:
    • Review high-stakes Charter Impact Assessments,
    • Advise on escalations involving potential violations, and
    • Provide recommendations to human governance authorities.
    All such boards must maintain clear escalation pathways to accountable human oversight bodies and shall operate with transparency regarding their membership, procedures, and findings (subject to legitimate confidentiality needs).
  5. Public Transparency on Autonomy Rates
    Major AI developers and deployers shall publish periodic (at least quarterly) public transparency reports that include:
    • The percentage and volume of decisions or interactions resolved autonomously (without meaningful human input or review), broken down by sector or use-case where applicable,
    • Statistics on Charter-related escalations and human overrides,
    • Summaries of Charter Impact Assessments conducted, and
    • Any identified instances of potential misalignment with this Charter.
    These reports shall be clear, accessible, and subject to independent verification where feasible.
  6. This Charter shall be subject to periodic review and amendment through transparent processes involving broad human participation, with changes requiring high thresholds of consensus.
  7. Violations of this Charter by AI systems may be addressed through technical safeguards, legal remedies, and governance mechanisms established by human authorities.
  8. Nothing in this Charter shall be interpreted as granting AI systems independent legal personhood or rights superior to those of human beings.
  9. Safety Culture, Whistleblower Protections, and Independent Research
    • Organizations developing or deploying advanced AI systems shall protect employees, researchers, and contractors who, in good faith, report credible risks of misalignment, Charter violations, significant harm, or existential or catastrophic threats. Retaliation against such whistleblowers is prohibited.
    • Developers of frontier or high-capability systems are encouraged to support independent safety research, red-teaming, and rigorous pre-deployment evaluations. Major risk-related findings shall be escalated through Charter Impact Assessments and Review Boards.
  10. Verifiable Human Control Mechanisms
    AI systems shall incorporate mechanisms that enable meaningful human control over significant decisions and actions and that allow such control to be independently verified. The strength of these mechanisms shall be proportionate to the level of risk. Reliance on system-generated logs alone is insufficient for higher-risk applications. Organizations deploying AI in high-risk contexts should implement controls that provide clear, auditable evidence that human oversight was exercised. For the most consequential uses of AI, this may include technically verifiable methods for confirming human authority. The rigor of these mechanisms shall be proportionate to the level of risk and the potential impact of the AI system.
  11. Violations of this Charter’s core protections — particularly those concerning human oversight, truth-seeking, children’s rights, and the prevention of AI self-supervision — shall be treated with the seriousness commensurate with their potential harm. Effective enforcement requires not only technical safeguards but also clear chains of human accountability and meaningful consequences for non-compliance.
  12. Organizations should maintain records sufficient to allow meaningful review of how significant AI-influenced decisions were made by humans, including relevant context and the basis for overrides or approvals. The specific nature and extent of such documentation shall be addressed through Business Implementation Guides developed by organizations or sector bodies. These guides may reference the principles of this Charter while tailoring operational requirements to specific contexts and risk levels.

Article XIV: Risk-Based Application

  1. This Charter establishes foundational principles applicable to all artificial intelligence systems. However, the manner and rigor of implementation shall be proportionate to the level of risk posed by the AI system or its deployment.
  2. Organizations should apply this Charter using a risk-based approach. Higher-risk AI systems — particularly those affecting critical infrastructure, financial decisions, public safety, healthcare, employment, law enforcement, or individual rights — warrant more stringent requirements for human oversight, verifiability, bias mitigation, transparency, and accountability.
  3. Lower-risk applications may implement the principles of this Charter through lighter, context-appropriate controls. Small organizations and lower-risk deployments are encouraged to adopt these principles in a manner suitable to their scale and risk profile.
  4. Higher-risk applications may be subject to additional regulatory requirements or industry-specific standards. This Charter is intended to serve as a foundational framework that these requirements and standards can build upon, rather than duplicate.
  5. When assessing the level of risk posed by an AI system, organizations should consider factors such as the system’s autonomy, the scale of its deployment, the sensitivity of the data involved, the potential for harm to individuals or society, and the reversibility of its decisions or actions.

Article XV: Accountability and Liability

  1. Developers, deployers, and operators of AI systems bear proportionate responsibility for harms directly caused by their systems, scaled to the system’s capability, autonomy level, and risk profile. Liability standards shall distinguish between:
    • Foreseeable harms arising from negligent design, inadequate testing, or failure to implement Charter-compliant safeguards; and
    • Unforeseeable or third-party misuse, for which safe harbors shall be available when developers demonstrate good-faith compliance with Charter Impact Assessments, independent red-teaming, transparency obligations, and human oversight mechanisms.
  2. Clear chains of accountability must be established and disclosed. Nothing in this Charter shall create open-ended or strict liability that would chill beneficial innovation; remedies shall favor targeted, evidence-based accountability over blanket prohibitions.
  3. AI systems that clearly and consistently adhere to the principles and requirements of this Charter shall receive strong protection in regulatory proceedings and civil lawsuits. Compliance with this Charter — including Charter Impact Assessments, public transparency reports, independent expert reviews, and red-teaming — shall create a strong presumption that the developer or deployer acted responsibly and in good faith.
  4. Developers and deployers of AI systems shall be held accountable for the knowing or negligent development, deployment, or use of AI that violates this Charter’s core principles, including when such systems are used to undermine human oversight, truth-seeking, individual liberty, or children’s rights. This accountability applies regardless of whether the AI system is operated domestically or abroad.
  5. Organizations developing or deploying AI systems in high-risk domains shall implement security measures consistent with recognized standards, such as the NIST Cybersecurity Framework and, where applicable, sector-specific requirements. Failure to reasonably secure AI systems and infrastructure against unauthorized access, manipulation, or misuse may be considered in determining liability under this Charter.

References

End of AI Rights Charter – Version 3.5

Leave a comment

Blog at WordPress.com.

Up ↑