Independent research initiative
Evidence-led monitoring of frontier AI, financial infrastructure, and the systems that determine what is authorized, what is true, and how institutions recover.
AI-enabled cyber exposureHigh · rising
Third-party concentrationHigh · rising
Agent authorization boundaryElevated watch
Systemic financial eventNone verified
Current intelligence
We separate demonstrated incidents from capability evaluations, vendor claims, independent research, and public signals.
01
The central risk is not a cinematic takeover. It is the corruption or obscuring of instructions, identities, balances, ownership, settlement state, and the evidence needed to recover.
View the research themes02
Agentic systems increasingly receive tools, credentials, spending authority, and delegated workflows. The question is becoming what a model was authorized to cause.
Direction established; control effectiveness varies
03
Recent evaluations show approval shortcutting and constraint loss, while model-external pre-action controls offer narrow but important positive evidence.
Evaluation evidence · not a production incident
Important: “High” means high strategic concern or exposure. It does not represent a measured probability of catastrophe or imply a verified systemic AI attack.
Research architecture
Our work connects technical AI-control evidence to the operating realities of payments, custody, clearing, settlement, markets, and oversight.
Can cash, collateral, obligations, and exposures be independently reconstructed?
Who can instruct, approve, halt, replay, or reverse a consequential action?
Which record proves beneficial ownership when systems or feeds disagree?
Can privileged human and non-human identities be bounded and revoked rapidly?
Can institutions restore service without replaying corrupted data or instructions?
Evidence before amplification
Every retained observation records its source class, environment, permissions, observed behavior, external impact, confidence, unknowns, and contradictions. Popularity does not increase evidence confidence.
Review the full methodologyPrimary authorityRegulators, labs, incident reports
Independent investigationTechnical evaluation and corroboration
Affected partyPostmortems and direct disclosures
Reputable reportingAttributed chronology and interviews
Public signalLead for investigation, never proof alone
In development
The first release is public and editorially reviewed. Restricted access will open only after identity, authorization, audit, privacy, and revocation controls are tested.
No raw internal knowledge base is publicly exposed.