M2MINT™ Research
Timeline
This timeline documents the development of M2MINT™, from its initial public research releases to subsequent conceptual materials and planned research activities.
Feb 2026
INITIAL RESEARCH AND CONCEPT DEVELOPMENT
Ziya Gökalp began the research that would lead to M2MINT™, exploring how autonomous and semi-autonomous AI agents might function as intelligence actors and intelligence targets, and how intelligence and counterintelligence relationships could develop between them.
Sep 2026
PUBLIC RELEASE OF THE FRAMEWORK
The English and Turkish preprints introduced Machine-to-Machine Intelligence (M2MINT™), a conceptual framework developed by Ziya Gökalp for examining intelligence and counterintelligence relationships among autonomous and semi-autonomous artificial agents.
Sep 2026
LAUNCH OF THE M2MINT™ WEBSITE
The launch of m2mint.org brought together the framework’s core concepts, research publications, governance considerations, and proposed directions for further investigation.
Sep 2026
HYPOTHETICAL SCENARIO AND BOUNDARY TEST
A hypothetical agent-to-agent scenario and the M2MINT™ Boundary Test were developed to illustrate the framework and clarify the distinction between routine interaction, authorized testing, systematic characterization, and adversarial collection. These materials are conceptual illustrations, not experimental results.
ONGOING - ACADEMIC EVALUATION
A manuscript presenting the target-centric M2MINT™ framework has been submitted to an academic journal for consideration. Submission does not imply acceptance or publication.
PLANNED - CONTROLLED EMPIRICAL INVESTIGATION
Future work will explore controlled experimental designs to assess behavioural prediction, the detection of systematic characterization, and the effects of counterintelligence responses.
FUTURE DEVELOPMENT - REVIEW INFORMED REFINEMENT
Further refinement of the framework will be guided by academic feedback, relevant literature, and empirical findings as they become available.
Future Roadmap
THEORETICAL REFINEMENT
Refine the framework’s concepts and boundaries through academic feedback, relevant literature, and critical evaluation.
EMPIRICAL VALIDATION
Develop controlled experiments to test behavioural prediction, the detection of systematic characterization, and the effectiveness of counterintelligence responses.
PRACTICAL APPLICATIONS
Explore how M2MINT™ could inform AI agent security assessments, governance practices, and tools for evaluating intelligence exposure.