Giskard vs Harmonic Security

Giskard is a Paris-based company founded in 2021 by Alex Combessie, Jean-Marie John-Mathews and Andrei Avtomonov. Combessie and John-Mathews are now co-CEOs, and Giskard says its co-founders previously built AI systems at Dataiku and Thales. It sells software and services for testing and securing LLM agents. Its first funding round, €1.5 million led by Elaia with Bessemer Venture Partners, was announced in December 2022. Giskard says it also received a €3 million strategic investment through the European Commission's EIC Accelerator in 2023 and a €3 million Bpifrance grant in 2024 for LLM evaluation research with Mistral AI, Artefact, INA and BnF. CORDIS lists a €2.5 million EU contribution to its EIC project, which ran from September 2023 to August 2025. Giskard Hub is the paid platform. Continuous red teaming runs multi-turn attacks that adapt to the agent's answers, generated from the company's own documents and knowledge bases. It tests for hallucinations, stereotypes and discrimination, harmful content, disclosure of personal information and prompt injection. Testing is black-box: the agent only needs an API endpoint, and coverage is limited to text-to-text conversational agents. The Hub also schedules agent evaluations and manages test datasets, custom checks and human review, with alerts, SSO, role-based access and audit logs. It runs as SaaS, in a private cloud or on premises, with EU or US data processing. Giskard Guards, launched in April 2026, is a runtime guardrail layer that runs in the customer's infrastructure. It inspects prompts, responses, tool calls and multi-step reasoning, and its policies are written as code in OPA/Rego. Giskard also sells AI security audits that end in a go/no-go report, and it is offering a free trial of a new Agentic Red Teamer to a limited number of users. The open-source library is Apache 2.0 licensed. Version 3, a rewrite for multi-turn agent testing, became stable in August 2026. Version 2 is no longer actively maintained, and tabular models have no v3 equivalent. Giskard also publishes Phare, an LLM safety benchmark launched in February 2025 with Google DeepMind as research partner. Logos on its site include BNP Paribas, AXA, BPCE, Société Générale, Crédit Agricole, Michelin, Decathlon, L'Oréal and Doctolib. **AI compliance coverage** Guards includes an EU AI Act policy pack for prohibited practices such as social scoring and real-time biometric identification, and Giskard says Guards covers OWASP LLM risks. It doesn't map its products to the NIST AI RMF or ISO 42001. Its trust center, run on Bastion, lists GDPR and a SOC 2 Type 2 report and shows HIPAA as in progress. **Who it's for** Enterprises, especially in European banking and insurance, that run customer-facing LLM agents and want testing and runtime controls they can deploy on premises.

Harmonic Security sells software that shows security teams how employees and AI agents use generative AI tools and stops sensitive data from going into them. It was founded in 2023 (its About page dates the founding to August 2023) by CEO Alastair Paterson, the former chief executive of Digital Shadows, and CTO Bryan Woolgar-O'Neil. It launched in October 2023, jointly headquartered in London and San Francisco, and its privacy policy names Harmonic Security Limited in London and Harmonic Security Inc. in San Francisco. It says it has raised $76 million in total: a $7 million seed round led by Ten Eleven Ventures (October 2023), a $17.5 million Series A led by Next47 (October 2024) and a $50 million Series B led by Crosspoint Capital Partners, announced on October 14, 2026, when it reported more than 100 customers and more than 90 staff in the US and UK. We found no acquisition. The platform is sold as three packages. Explore uses a browser extension to find the AI tools in use and gives prompt-level visibility, usage intelligence, sensitive data detection and risk scores for each app; Harmonic says it covers more than 1,000 AI applications, including AI features inside SaaS tools. Guide adds an endpoint agent for desktop AI apps, IDEs and command-line tools such as Claude Code and Codex, inline coaching (block, warn or log), granular policies and webhook alerts for SIEM and SOAR tools. Command adds MCP client and server discovery, permissions on what agents may read, write and execute, and data protection inside agentic workflows; prompt injection detection is listed as coming soon. Its locally installed MCP Gateway, launched in October 2025, intercepts MCP traffic on Windows, macOS and Linux. Instead of pattern matching, Harmonic classifies prompts with its own small language models, which it says decide in under 200 milliseconds. **AI compliance coverage** Harmonic treats compliance as a by-product of governance. Its compliance page says it flags high-risk uses such as automated hiring screens and credit scoring "so you can apply the specific guardrails the EU AI Act and local laws require", keeps an audit trail of AI interactions, and blocks health and personal data from unapproved AI tools for HIPAA and GDPR. Documentation mapping its controls to specific regulations is available on request, as is EU hosting. For Harmonic itself, its trust center lists SOC 2 Type 2 and ISO/IEC 27001:2022, with a SOC 2 report available on request, and its site footer shows SOC 2 Type 2 and ISO 27001 badges. ISO/IEC 42001 and the NIST AI RMF also appear in the trust center's list, but the site shows no badge or certification claim for either. **Who it's for** Security and IT teams that want to let staff use ChatGPT, Claude, Copilot, Cursor and similar tools while keeping company data out of them. Harmonic has industry pages for technology, legal, insurance and healthcare, and its site quotes customers at Apax and Advisor360.