Digital Colonization and Consent

The risks of digital colonization are not hypothetical. They are happening now. Data about Indigenous peoples is being collected, aggregated, and fed into AI systems without meaningful consent. Contracts transfer tribal data to outside platforms that then charge tribes for access to their own information. Publicly available Indigenous knowledge, language, and cultural materials are used to train commercial AI models with no tribal involvement, no compensation, and no mechanism for tribes to object.

This section names these risks directly and frames Free, Prior, and Informed Consent as the governing standard for how tribes engage with AI and digital systems.

Data Extraction Without Consent

AI systems are trained on vast quantities of data scraped from the public internet and from datasets assembled by commercial and academic entities. Indigenous cultural materials, language resources, historical records, and community information are routinely included in these datasets without tribal knowledge or approval. The result is that AI models learn patterns from Indigenous data and generate outputs shaped by it, while tribes have no visibility into or control over that process.

This is not a privacy problem in the conventional sense. It is a sovereignty problem. When a commercial AI company trains a model on Cherokee language resources, Navajo textile patterns, or Lakota oral histories available online, it is extracting value from sovereign cultural assets without the consent of the nations to whom those assets belong.

Contracts That Transfer Data

Many technology contracts, particularly for cloud services and software-as-a-service platforms, include terms that grant the vendor broad rights over data processed through their systems. In some cases, this means tribal enrollment records, health data, educational information, or economic development data is stored on outside servers, subject to outside jurisdiction, and accessible to the vendor for purposes including product improvement, analytics, or model training.

Worse, some vendors then charge tribes for access to their own data, or make data portability so difficult that switching providers becomes prohibitively expensive. This creates a dependency relationship that mirrors historical patterns of resource extraction: the resource (data) is removed from tribal control, processed elsewhere for outside benefit, and sold back to the tribe at a markup.

The Question of AI and Assimilation

A deeper concern surfaced at the Tribal Innovation Summit: what if AI systems carry an inherent bias shaped by the colonial worldview embedded in their training data? If the vast majority of an AI model’s training corpus reflects Western epistemologies, legal frameworks, and cultural assumptions, then the model’s outputs will reflect those assumptions. When tribes use such a model for governance, education, or community services, they may be quietly adopting a framework that does not align with their values, governance structures, or ways of knowing.

This is not a call to reject AI. It is a call to use it with clear eyes and strong governance. The question is not whether AI is useful. The question is whether tribes can shape how it is deployed so that it serves tribal priorities rather than subtly rewriting them.

Free, Prior, and Informed Consent as a Governing Standard

Free, Prior, and Informed Consent is the foundational principle of international Indigenous rights law that recognizes the right of Indigenous peoples to make autonomous decisions regarding activities that may affect their lands, resources, cultures, or communities. FPIC is articulated most fully in the United Nations Declaration on the Rights of Indigenous Peoples, which affirms that consent must be free from coercion or manipulation, given prior to the commencement of activities, based on full disclosure of relevant information, and grounded in meaningful Indigenous participation in decision-making.¹

Although FPIC originally developed in the context of land rights and natural resource extraction, it applies with equal force to the digital domain.² In the modern information economy, Indigenous data functions as a valuable resource capable of being collected, analyzed, commercialized, and integrated into AI systems. This resource should be governed by the same consent-based principles that apply to Indigenous lands and natural resources.

The Playbook’s approach to FPIC is developed in detail in the Execution phase, which includes a tiered consent model, implementation guides, and practical templates. The core principle established here in Module 0 is that FPIC is a minimum standard, not a checkbox. Consent must be ongoing, revocable, and accompanied by genuine tribal authority over the terms of engagement.

Open Legal and Policy Questions

Several critical questions remain unresolved and represent active areas of development for the Playbook and the broader Open Sovereignty Lab community:

Can tribes legally block AI companies from using publicly available tribal data for model training? What mechanisms exist, and what new ones need to be created?

Can tribes charge for the use of their data in AI training? What would a licensing or compensation framework look like, and how would it be enforced?

How should tribes approach AI systems that have already been trained on their data without consent? Is there a remedy, a renegotiation, or a governance response?

What does “informed” consent mean when the technology itself is opaque? If a vendor cannot explain how an AI model uses tribal data internally, can consent be considered informed?

These are not questions with settled answers. They are the frontier of tribal technology policy, and the Open Sovereignty Lab exists in part to convene the expertise needed to address them.

---

Legal Foundations: FPIC in International and Digital Contexts

Free, Prior, and Informed Consent (FPIC) is a foundational principle of international Indigenous rights law that recognizes the right of Indigenous peoples to make autonomous decisions regarding activities that may affect their lands, resources, cultures, or communities. FPIC is most prominently articulated in the United Nations Declaration on the Rights of Indigenous Peoples (UNDRIP), which affirms that Indigenous peoples must be consulted and provide consent before states, corporations, researchers, or other entities undertake actions that impact their rights or interests.¹ The principle is composed of four interconnected elements: consent must be “free” from coercion or manipulation, “prior” to the commencement of activities, “informed” through full disclosure of relevant information, and based on meaningful Indigenous participation in decision-making processes.

Although FPIC originally developed in the context of land rights and natural resource extraction, it has become increasingly important in discussions surrounding Indigenous data sovereignty and digital governance.² In the modern information economy, Indigenous data functions as a valuable resource capable of being collected, analyzed, commercialized, and integrated into artificial intelligence systems. Therefore, such data should be governed by the same consent-based principles that apply to Indigenous lands and resources.

The application of FPIC to AI and digital systems raises new implementation questions that the existing international framework does not fully address. Traditional FPIC processes assume a discrete project or activity for which consent can be sought at a defined point in time. AI systems, by contrast, involve continuous data processing, model updates, and evolving outputs. Consent in this context must be understood as ongoing rather than one-time, revocable rather than permanent, and accompanied by mechanisms for tribes to monitor and audit how their data is being used.

The tiered consent model discussed at the 2026 Tribal Innovation Summit offers one practical approach: basic queries receive agnostic, non-personalized responses, while tailored responses that draw on tribal-specific data require explicit opt-in consent with clear disclosure of how data will be used and why. This model preserves tribal choice at every level of engagement and ensures that consent is not assumed from mere use of a platform.

---

References

Footnotes

1. United Nations Declaration on the Rights of Indigenous Peoples (UNDRIP); see also FAO, “Free, Prior, and Informed Consent: An Indigenous Peoples’ Right and a Good Practice for Local Communities” (2016). https://www.un.org/development/desa/indigenouspeoples/publications/2016/10/free-prior-and-informed-consent-an-indigenous-peoples-right-and-a-good-practice-for-local-communities-fao/ ↩ ↩²

2. See IHRB, “What is Free, Prior, and Informed Consent?” https://www.ihrb.org/resources/what-is-free-prior-and-informed-consent-fpic ↩ ↩²