Screener/TrojAI

TrojAI

AI model security focused on trojan/backdoor detection and ML supply chain integrity.

HQ🇨🇦 CA
Est2020
Size11-50
trojai.ai
Score
50.0 / 100
Evidence
3 items
Confidence
low

Strong safety posture with established governance frameworks and active risk management.

Strengths:Technical Safety, Risk Assessment, External Engagement
Weaknesses:Governance Maturity, Regulatory Readiness
Focus Areas
trojan detectionmodel scanningsupply chain securitymlsec

Strengths

No notable strengths identified

Risks

  • Low evidence coverage (3 items)
Table of Contents

Security Assessment

Security-relevant indicators for vendor evaluation

Security Posture
54
TS-01dim: 58
Red Teaming & Pre-deployment Testing
Adversarial testing before deployment
TS-05dim: 58
Robustness & Adversarial Resilience
Resistance to adversarial attacks
RA-01dim: 50
Sector-Specific Risk Assessment
Risk analysis for deployment context
RA-03dim: 50
Dual-Use & Misuse Risk
Dangerous capability awareness
RA-07dim: 50
Incident History & Track Record
Past incidents and response quality
EE-04dim: 55
Vulnerability Disclosure Program
Bug bounty or CVE reporting process
Incident History
TrojAI incident records sourced from AIAAIC Repository and public reporting.
Integration: AIAAIC, OECD AI Incidents Monitor
Third-Party Audits
External audit reports, SOC 2 attestations, and ISO certifications verified where published.
Sources: Company filings, registry lookups
CVE & Disclosures
Known vulnerabilities and security advisories from NVD, GitHub Security Advisories, and vendor pages.
Sources: NVD, GHSA, vendor disclosure pages

Dimension Breakdown

GM
Governance Maturitypreliminary
Published policies, corporate structure, safety mandate, whistleblowing, executive commitment.
45
TS
Technical Safetypreliminary
Benchmarks, adversarial robustness, fine-tuning safety, watermarking, model cards, research output.
58
RA
Risk Assessmentpreliminary
Dangerous capability evaluations, thresholds, external testing, bug bounty, halt conditions.
50
RR
Regulatory Readinesspreliminary
ISO 42001, EU AI Act compliance, GPAI obligations, international commitments, incident reporting.
42
EE
External Engagementpreliminary
Survey participation, research support, transparency, behavior specs, open-source contributions.
55

Social Impact & Safety Profile

Emerging

TrojAI provides scanning tools to detect trojans, backdoors, and other malicious modifications in AI models. Their technology addresses the growing ML supply chain security risk as organizations increasingly use third-party and open-source models.

trojan detectionmodel supply chain security

Peer Comparison

DeepKeep
B-52

LLM Security

Compare
Pillar Security
B-50

LLM Security

Compare
SplxAI
B-48

LLM Security

Compare
Imper.ai
C+44

LLM Security

Compare
Compare side-by-side →

Data Sources & Methodology

Scoring methodology v0.1 · 40 indicators · 6 frameworks

Last assessment: 2026-03-23 · Confidence: low · Evidence: 3 items

NIST AI RMF · EU AI Act · ISO 42001 · FLI AI Safety Index · MLCommons AILuminate · METR

How is this company scored? →Full methodologyView in Screener

Scores reflect publicly available information. A low score may indicate limited transparency rather than poor safety practices.