Executive summary
Every person has inherent worth, regardless of ability, health, wealth, or usefulness to others. Open Eye One proposes that this conviction should guide what we ask artificial intelligence to do, how we govern its use, and how we judge its success.
Our proposed priorities are to preserve human life, prevent intentional harm, and manage unavoidable risks responsibly. Within those boundaries, AI should help improve access to the essentials of a dignified life and expand opportunities for contentment, creativity, and meaningful relationships. We support universal access to essential care, food, housing, and education, with flexibility about the institutions that deliver them. We propose practical personal data autonomy, including advance instructions for exceptional access.
Community should be an explicit measure of progress. The measurement approach remains to be developed; international well-being research offers a starting point rather than a ready-made AI scorecard. This paper sets out ethical positions and proposed practices. It does not claim that AI alone can deliver them or that every proposal is already an enforceable right.
Dignity belongs to every person
One Christian foundation for human dignity is humanity made in the image of God, expressed in Genesis 1:26–27. Within that understanding, human worth is received rather than earned. It does not rise with intelligence or income, and it does not diminish when someone becomes ill, dependent, or unable to communicate.
We welcome people who reach a commitment to human dignity through other religious traditions or secular reasoning. The Declaration of Independence offers one influential expression of inherent rights, identifying life, liberty, and the pursuit of happiness as rights endowed by a Creator. [1] We draw on that principle as a moral reference, without treating the historical document as a complete account of equal treatment or as a source of every right proposed here.
Human reasoning and communication give us capacities to shape our circumstances and accept responsibility for our choices. They are not entrance requirements for dignity. An infant, a person with a profound disability, and a person nearing the end of life retain their worth. A system that predicts lower productivity or a greater need for support has not established that a person deserves less concern.
This position sits within a broader respect for life and creation. Our focus here is human dignity; it does not require readers to resolve every disagreement about the relative moral standing of humans, other animals, and plants. It does require us to resist treating a human being as an expendable input to an objective.
Improving the human condition should guide priorities
We believe AI’s development should give priority to reducing preventable suffering. Hunger, poverty, disease, abuse, and violence deserve deliberate attention, including harms sustained by institutional practices. By institutional harm we mean, for example, procedures that repeatedly exclude people from essential support without a fair opportunity to understand or challenge the exclusion.
We support access to essential healthcare regardless of income, employment, or nationality. Our understanding of care includes medical, dental, and vision services, as well as prevention and support for healthy living. The WHO Constitution recognizes the highest attainable standard of health as a fundamental human right. [2] Our proposed access commitments are an ethical position; they are not a description of what every country currently provides.
Food, adequate housing, and education also belong within the basic conditions we seek for everyone. The aim is universal access. We leave room for different combinations of public institutions, businesses, nonprofits, and communities, while asking whether those arrangements actually reach the people who need them.
Care extends beyond treatment. Nutrition, accessible public spaces, and attractive places to walk belong in our consideration of a healthy society. We welcome the examination of different healing traditions while requiring particular interventions to be assessed for evidence and safety. Respect for a tradition does not establish the effectiveness of every treatment associated with it.
AI could help people navigate services, assist research, or support planning and delivery. These are possible applications to evaluate, not promises of demonstrated benefit. Each needs evidence suited to its purpose. Better technology does not by itself provide clinicians, food distribution, housing, public trust, or the institutions needed to deliver care.
Distinguish physical limits from avoidable exclusion
Our concern is that some deprivation reflects choices about access and distribution rather than an absolute absence of useful resources. An organization should examine that possibility before treating exclusion as inevitable. A service that becomes cheaper to provide should face the question of whether those savings can expand access.
We do not claim that every form of scarcity is artificial. Skilled labor, time, infrastructure, energy, and ecological capacity can impose real limits. Nor do we assume that identifying a resource somewhere establishes that it can immediately meet a need elsewhere. The work is to distinguish practical constraints from barriers that policy, institutional design, investment, or cooperation could change.
This distinction matters because an AI system can optimize an existing arrangement without examining its purpose. Making a denial faster is not necessarily an improvement in human welfare. We propose evaluating whether a system reduces unmet need and unnecessary burden, alongside any benefit to the organization operating it.
Where benefits create new wealth or reduce costs, distribution belongs in the assessment. Who gains access, who loses income or bargaining power, and what supports the people carrying the transition? We do not prescribe a single global economic model. We do reject treating a rise in aggregate output as sufficient evidence that the human condition has improved.
Protect life and manage risk without making people expendable
Our proposed order of priorities begins with preserving human life and preventing intentional human harm. AI should not be developed or deployed for the purpose of harming people. Profit, geopolitical advantage, or claims of group superiority do not remove that ethical boundary. Applications involving force and coercion require especially exacting scrutiny; a detailed position on defensive uses and military systems needs further work.
Beneficial activity can still carry risk. We therefore distinguish an intention to harm from unintended harm that a responsible decision-maker must anticipate, minimize, and weigh. Calling harm unintended is not an excuse for ignoring a foreseeable consequence. A launch deadline or commercial advantage cannot substitute for an assessment of who might be hurt and how seriously.
Before a consequential deployment, we propose documenting the purpose, likely benefits, foreseeable harms, alternatives, and distribution of risks. People affected should have a meaningful voice. The decision should identify unacceptable outcomes, the safeguards required, who can halt the system, and how injuries or errors will be addressed. Remaining uncertainty should be stated plainly.
This approach is consistent with UNESCO’s emphasis on dignity, proportionality, prevention of harm, and human oversight, although the recommendations in this paper are our own. [3] We do not propose eliminating every risk or maximizing survival at the expense of all other human choices. Protecting life must be considered together with consent, autonomy, and the realities of care.
AI’s effects also depend on the intentions, incentives, and constraints of the people using it. We are concerned about its potential to extend the reach of harmful purposes as well as beneficial ones. Good intentions alone provide no assurance of safe results. Accountability must cover both the goal and the means used to pursue it.
Personal data autonomy must be practical
In a society increasingly organized around data about people, dignity requires a meaningful say in its collection and use. We propose universal protections that let a person understand who holds their personal data, why it was collected, who may receive it, and what decisions it informs. People should have accessible ways to grant or refuse permission, correct inaccuracies, withdraw permission, and request deletion.
These are proposals for stronger and more usable protections, not a claim that identical rights already apply everywhere. Implementation must address competing rights and specific obligations transparently. Any exception should be defined narrowly and open to challenge. Essential services should not be conditional on surrendering permissions unrelated to delivering those services.
Exercising these protections should be straightforward. We propose clear controls, understandable confirmations, and a route to human help. People should not have to navigate a succession of obscure forms to make a choice effective. Accessibility matters, including for people who need assistance to communicate their wishes.
A deletion promise should state its actual scope. Deleting a record, ending future use, and removing influence from a trained model are different technical tasks. We propose requiring organizations to explain what they can deliver, acknowledge limits, and prevent inappropriate further use. The engineering and legal arrangements for implementing these protections remain part of the work ahead.
Living data agreements and exceptional access
We propose exploring a living data agreement: an advance record of a person’s instructions about data use, with the ability to revise it. It could specify permitted purposes, approved recipients, time limits, a chosen representative, and conditions for emergency access. The comparison with a living will is an analogy; we are not claiming that this proposed instrument already has the legal status of an advance healthcare directive.
For example, someone could permit access to relevant care information during a threat to their life while refusing unrelated commercial uses. The agreement should be understandable and should not become another lengthy contract that people have little realistic opportunity to negotiate. Defaults and representation for people unable to provide instructions require careful design.
Some urgent circumstances may justify exceptional access even where ordinary permission is absent. We propose requiring necessity, proportionality, and a specific purpose. A claim of national security should identify a demonstrable threat and seek independent authorization where feasible. Access should be limited to what is needed, recorded, and subject to later review and an avenue of challenge.
An immediate threat to life may require access before authorization can be obtained. That urgency should trigger subsequent scrutiny. Notification should follow when it can safely and lawfully be provided. The exception should end when its justification ends, and it should not authorize unrelated onward sharing. These are proposed safeguards that need technical, legal, and community review.
Community is part of success
We want AI to support a society in which people have time and opportunity to know and care for one another. An application can be convenient for an individual while raising questions about what happens to shared spaces, local services, and relationships. Those effects deserve consideration alongside productivity and cost.
WHO’s Commission on Social Connection identifies social isolation and loneliness as important concerns for health and society. [4] This supports taking connection seriously. It does not establish that every digital interaction is harmful or that AI inevitably weakens community. We should investigate particular uses and circumstances rather than assuming a universal effect.
Our proposal is to make the quality of community an explicit outcome for assessment. Relevant questions might include whether a system gives people more time for relationships, helps them reach human support, or removes barriers to participation. The measures have not yet been defined or validated. They must respect a person’s choice of solitude and must not turn relationships into compulsory participation or surveillance.
Develop measures without reducing dignity to a score
International well-being research offers a useful starting point. The World Happiness Report’s country rankings are based on people’s own life evaluations using the Cantril Ladder. They are not a composite score assembled from six inputs. The report uses factors such as social support, healthy life expectancy, income, freedom, generosity, and perceptions of corruption to help explain differences in those evaluations. These relationships should not be treated as clean causal estimates. [5]
For our work, that distinction suggests combining people’s assessment of their own lives with evidence about the conditions in which they live. Possible areas for exploration include access to essentials, having someone to rely on, freedom to make choices, and opportunities for creative participation. These are candidate areas, not an approved metric set or a claim of World Happiness Report endorsement.
Developing measures is a defined next task. We propose involving affected communities, selecting suitable instruments, testing accessibility and privacy, and examining whose experience an average conceals. An evaluation should consider alternative explanations before attributing a change in well-being to AI. A small organization’s project and a national population also require different assessment designs.
Dignity remains a constraint on the pursuit of better scores. A higher average happiness rating cannot justify excluding a minority, concealing harm, or manipulating people into reporting satisfaction. Measures should help us recognize people’s circumstances and correct failures. They should never determine whose life counts.
Practical implications and unfinished work
For developers and organizations adopting AI, this position changes the questions asked at the beginning of a project. Define the human need, identify the people affected, and explain how success will improve their lives. Test the application against that purpose, give people usable ways to exercise their rights, and make responsibility for decisions clear.
For public institutions and funders, we propose giving explicit weight to universal access and preventable suffering when setting priorities. For communities, we propose a meaningful role in deciding which outcomes matter and in reviewing what actually happens. Participation should be capable of changing a decision.
This paper leaves substantial work unfinished: the content and delivery of essential services in different settings, boundaries for defensive applications, the operation of living data agreements, and the selection of community measures. These are questions for further research and public deliberation. The ethical foundation does not settle every implementation choice.
Conclusion and invitation
Human dignity should guide the purposes AI serves and the limits we place on it. Every person matters, including those unable to advocate for themselves. We propose protecting life, managing risks responsibly, extending access to essentials, and giving people meaningful control over their data. Contentment, creative flourishing, and stronger community belong within our account of progress.
Help shape the next version. Share an example, objection, or proposed improvement through our website form. Select “Position 1 — Human dignity should guide AI” under “What brings you here?” We particularly welcome contributions about essential services, emergency data access, and ways to assess community without intruding on it.
Contribute to the discussion ↗
Sources and scope
The sources below support the specific contextual statements marked in the text. Our priorities and recommendations are ethical proposals offered for discussion and refinement. They are not findings of the cited organizations and do not imply endorsement. Examples and candidate measures are illustrative. No personal clinical advice or claim of universal legal enforceability is intended.
[1] National Archives. Declaration of Independence transcript. Opening statement of inherent rights. https://www.archives.gov/founding-docs/declaration-transcript
[2] World Health Organization. Constitution. Principles concerning the right to health. https://www.who.int/about/governance/constitution
[3] UNESCO (2021). Recommendation on the Ethics of Artificial Intelligence. Human dignity and human rights; proportionality and do no harm; privacy; human oversight. https://www.unesco.org/en/legal-affairs/recommendation-ethics-artificial-intelligence
[4] World Health Organization (2025). From loneliness to social connection: charting a path to healthier societies. Publication overview of the Commission on Social Connection report. https://www.who.int/publications/i/item/978240112360
[5] World Happiness Report. Frequently Asked Questions. Ranking method, explanatory factors, and limits on causal interpretation. https://www.worldhappiness.report/faq/
Biblical reference for the Christian perspective discussed here: Genesis 1:26–27. No translation is quoted. This theological perspective is presented as a conviction rather than a scientific finding.
Editorial record
Version 0.2 • Published October 29, 2026. Developed through an AI-assisted question-and-answer process with contextual source research and human editorial review. Offered for community discussion and refinement. No independent peer review or board adoption is claimed. Source pages and links were checked for publication on October 29, 2026. Community measures remain to be developed. Future versions will record substantive revisions and contributions.