Dimension scores are derived from public data and fields; weighted into the composite. Reference only.
Junyul is an evidence platform for enterprise AI operational risk. Its core purpose is not to replace legal advice, but to connect AI assets, impact assessments, runtime events, incident investigations, reporting, and audit evidence. The website focuses on requirements related to Korea’s AI Basic Act, the EU AI Act, NIST AI RMF, OWASP Agentic Top 10, and Korean personal information and credit information regulations.
The product covers the Discover, Assess, Instrument, Investigate, Report, and Audit workflows. It can register AI models, Agents, and automated decision-making services, track owners, risk levels, and SDK integration status; perform structured impact assessments for high-impact AI, automated decision-making, personal information, and credit information risks; and record events such as inference, tool calls, retrieval, and automated decisions via Python, TypeScript, and Go SDKs. On the investigation side, it provides an Incident Command Dashboard that uses timelines, impact scope, forensic evidence, and root-cause analysis to support scenarios such as prompt injection, tool misuse, privilege abuse, suspected data leakage, and similar incidents.
Junyul’s most distinctive design choice is hash_only by default: raw prompts, responses, and tool parameters are not sent to Junyul servers. The SDK generates SHA-256 hashes locally and uploads only evidence fingerprints and minimal metadata. Forensic mode requires customer-managed keys. The architecture mentions a Seoul region, TLS, schema validation, rate limits, hash chains, Merkle roots, BYOK, and related controls. Integrations include API Key, Webhook, Slack, SIEM, customer KMS, red-team workflows, and OpenTelemetry. Deployment appears to be cloud-service oriented, but self-hosting is not specified.
The website does not publish plans or pricing, stating only that pricing and contracts are determined by the scope of organizational onboarding. The SDK can be installed publicly, but production use is controlled by a workspace API Key and paid contract. Its strengths are a clear positioning, detailed compliance evidence-chain design, and suitability for joint participation by security and legal teams. The drawbacks are that it does not disclose SLA details, permission granularity, real customer cases, or maturity metrics. It also repeatedly uses wording such as “designed to” and “implementation review,” so a PoC and security review are necessary before procurement.
Junyul is better suited to companies operating AI Agents, RAG systems, automated approval workflows, or high-impact AI in Korea while also paying attention to the EU AI Act. Information on access and payment from China is unknown. If the main focus is China-local compliance, MLPS, data export controls, or Chinese ecosystem integrations, organizations should also evaluate local AI governance, log auditing, SIEM, GRC, and model security platforms as alternatives or complements.
⚠ This review is compiled from public sources and does not constitute a purchase recommendation. Verify all facts on the vendor's official site. Verify on junyul.com official site.
junyul.com is an South Korea SaaS provider. TG4G tracks its product information, an overall rating of 7.0/10, and a China-accessibility score of Workable. Click "Visit Official Site" to reach junyul.com directly.