Credo AI vs Lasso Security

Credo AI sells an AI governance platform for enterprises. The company was founded in 2020 by Navrina Singh (CEO) and Eli Chen and is based in Palo Alto, California. It raised $21 million in July 2024 from new investors CrimsoNox Capital, Mozilla Ventures and FPV Ventures and existing investors Sands Capital, Decibel VC, Booz Allen Hamilton and AI Fund, which took total funding to $41.3 million. Customers named on its site include Mastercard, Northrop Grumman, Booz Allen, Principal, Autodesk, Madrigal Pharmaceuticals and AdeptID. The platform has three main parts. The AI Registry catalogs AI systems, models, agents and AI vendors, with shadow AI discovery, risk classification, dependency mapping and approval workflows. Risk Intelligence assesses each use case against a library of risks and controls, including agent risks such as tool misuse and scope drift, and takes monitoring signals from connected observability tools. The Policy Engine turns policies into controls using pre-built policy packs and records evidence for audits. Credo AI's homepage names the EU AI Act, NIST AI RMF, ISO 42001, OMB M-25, CO ADMT (Colorado) and NAIC AI; its regulations page also names NYC Local Law 144, Colorado SB 21-169 and SOC 2. Recent additions: the AI Agent Registry has been in public preview since September 24, 2025, and Credo AI still invites buyers to join the preview. A separate Agent Governor is labeled a research preview. GAIA (Govern AI Assistant), which drafts use case intake, suggests questionnaire answers and recommends risks and controls, has been generally available to all platform customers since May 13, 2026. Credo AI is sold as SaaS, including through AWS Marketplace, and it also documents a self-hosted install in the customer's private cloud on an existing Kubernetes cluster or a virtual machine. The company says it has a SOC 2 Type II report covering security, availability and confidentiality and that it repeats the audit every year. Credo AI is a governance layer, not a testing tool. Its current product pages describe no bias or fairness test, and its open-source Lens assessment library has been archived on GitHub since July 12, 2024. Forrester named Credo AI a Leader in The Forrester Wave: AI Governance Solutions, Q3 2025.

Lasso Security is an AI security company founded in 2023, with offices in Tel Aviv and New York. Its 2023 launch release names four founders: Elad Schulman (CEO), Ophir Dror (now CPO), Lior Ziv and Yuval Abadi. It came out of stealth on November 20, 2023 with a $6 million seed round led by Entrée Capital, with Samsung Next. On September 2, 2026 it announced a $30 million round led by ClearSky, with Entrée Capital, iAngels, Singtel Innov8, Mindset and Swish Data; SiliconANGLE put the total raised at more than $37 million. We found no acquisition of Lasso as of November 11, 2026. **What it sells** Lasso describes one platform with five parts. Discovery & AI-BOM inventories agents and applications and maps their models, prompts, tools and guardrails. AI Security Posture Management reviews misconfigurations, policy gaps and supply chain risk. Automated AI Red Teaming runs adversarial tests, including multi-turn agentic attacks and context poisoning, from a library Lasso puts at 3,000+ attacks. Runtime Enforcement applies policies inline at the proxy, API or AI gateway layer. AI Detection & Response detects threats mapped to MITRE and OWASP and uses intent analysis to spot behavioral anomalies. AI Usage Control covers employee use of AI tools: Lasso says it discovers shadow AI across more than 8,000 AI tools, detects sensitive data such as PII and payment card data in prompts and file uploads, and masks it before it reaches third-party models. With the September 2026 round Lasso announced two guardrail engines. LEAP is a transformer-free guardrail that runs on ordinary CPUs without a GPU; Lasso says it returns a verdict in under five milliseconds. RAPID is a self-hosted LLM-as-a-judge for decisions that need a plain-language policy; a routing layer sends most traffic to LEAP. These speed and cost figures are Lasso's own benchmarks. Its pages also cite an accuracy figure for its detections, but the homepage and its site index give different numbers (98.6% and 99.8%), so we don't repeat one. **Deployment and integrations** Lasso lists deployment through a reverse proxy, SDKs, a browser extension and API-level monitoring, and says LEAP can run in customer clouds, regulated environments and air-gapped networks. Its site index lists integrations with Palo Alto Networks, CrowdStrike Falcon, Okta, Cloudflare, Vertex AI, AWS Bedrock, Microsoft Azure and Salesforce Agentforce. In July 2025 it formed Lasso Federal LLC with Swish Data and said its products are available to US federal customers through AWS GovCloud (US). That release says agencies can deploy Lasso while meeting requirements such as FedRAMP High, DoD SRG, ITAR and CJIS; it does not say Lasso holds a FedRAMP authorization. **Frameworks and attestations** Lasso says policies can be mapped to NIST, OWASP and MITRE using templates and that it generates audit-ready reports. Its AI Usage Control page lists the NIST AI RMF, EU AI Act, GDPR, SOC 2 Type 2, ISO 27001, ISO 42001, HIPAA, PCI DSS, FedRAMP and the OWASP Top 10 for LLM Applications as frameworks it supports. A blog post says its policies can be aligned with the clauses of ISO/IEC 42001, without a clause-by-clause mapping. Lasso says the company itself is SOC 2 Type II certified. **Recognition and customers** Lasso cites a Gartner Cool Vendor mention (2024) and being named a Representative Vendor in Gartner's Market Guide for AI TRiSM, 2025. It is a member of NIST's AI Safety Institute Consortium. Customers it names include the US Department of Homeland Security, eToro, Fiverr, Kaufman Rossin and BMW. It publishes no prices.