AI research

DOSSIERVEILLANCE
DOSSIERVEILLANCE is an independent research, consulting, and policy services provider founded in 2019 by Cristina Plamadeala, PhD. DOSSIERVEILLANCE works with institutions, policymakers, and researchers to examine how historical and contemporary surveillance practices shape power, accountability, and human rights.

Through comparative and interdisciplinary analysis, DOSSIERVEILLANCE connects surveillance under communism with present-day institutional monitoring, data collection, and emerging AI technologies.
Our Framework
We have a structured methodology in our consulting, advisory, and educational services to help organizations design, evaluate, and govern AI-assisted systems responsibly.
This framework is built around a small set of strict principles intended to reduce fear, bias, and mistrust in automated decision-making while strengthening transparency, accountability, and human oversight.
Dossiers as the Operational Core
Every AI-assisted process is anchored in a dossier, consolidating data, AI outputs, human analysis, and decisions into a traceable record. Dossiers remain open to revision for reflection and correction.
AI as an Investigative Instrument
AI systems serve as investigative tools to surface signals and identify patterns. Final judgments and accountability remain with human decision-makers, recorded within the dossier.
Preserving Uncertainty
Uncertainty, limitations, and contested interpretations are explicitly preserved and documented. This prevents false certainty and reduces risks of biased automated outcomes.
Traceable Responsibility
Responsibility is always traceable. The framework records AI contributions, human interventions, and decision authority at each stage, ensuring accountability is never obscured by automation.
Investigable Decisions
All decisions are investigable. The dossier allows reconstruction of conclusions, alternatives, and trade-offs, enabling meaningful review, correction, and continuous learning.
Behavioral Effects as Design Risks
Behavioral impacts like fear, self-surveillance, and forced conformity are treated as system-level risks. Systems using opacity or intimidation are considered design failures.
Continuous Learning and Improvement
Learning is continuous. Dossiers function as living records that evolve as new information emerges, preserving past errors as resources for improving long-term AI deployment.
THE PROBLEMS WITH AI TECHNOLOGIES THAT WE SEEK TO SOLVE
DOSSIERVEILLANCE works with organizations, researchers, and institutions confronting a central problem in contemporary AI systems: these systems produce answers without delivering understanding.
Modern AI technologies generate classifications, predictions, and rankings with remarkable speed and confidence. Yet they rarely provide clear explanations for how decisions are made or how conclusions should be interpreted in real-world contexts. Automated systems now influence critical outcomes—loan approvals, hiring decisions, content moderation, criminal justice assessments—while remaining largely opaque to those affected by them.

Opacity, Risk, and Unequal Impact
As AI systems expand across sectors, their lack of transparency creates growing concern. The inability to meaningfully question or interpret automated judgments introduces structural risk, particularly where biased data, historical inequities, or surveillance practices are involved.
Employment
AI systems influence hiring, performance reviews, and career progression, raising questions of fairness and bias.
Healthcare
Algorithmic decisions in diagnosis and treatment impact patient care, necessitating transparency and accountability.
Finance
Loan approvals, credit scores, and fraud detection rely on AI, which can perpetuate financial inequities.
Public Governance
From social services to policing, AI in government requires scrutiny to prevent disproportionate effects on communities.
DOSSIERVEILLANCE examines how opacity in automated decision-making can reinforce discrimination, amplify institutional fear, and undermine accountability. These effects are not evenly distributed; they disproportionately affect individuals and communities subject to heightened monitoring and evaluation.

Fear, Anxiety, and Behavioral Consequences
Being evaluated by systems that cannot be fully understood or challenged produces rational anxiety. Over time, this anxiety alters behavior—within workplaces, public institutions, and society at large—reshaping trust, compliance, and decision-making norms.
DOSSIERVEILLANCE helps organizations understand how algorithmic surveillance and opaque AI systems influence human behavior, institutional culture, and long-term governance outcomes.
Services Offered by DOSSIERVEILLANCE
Consulting services on AI transparency, surveillance risk, and algorithmic bias
Educational workshops, seminars, and briefings for professionals and institutions
Research, analytical reports, and published writing on AI governance and accountability
Advisory services for organizations deploying automated or data-driven decision systems