SydeLabs

SydeLabs operates in the AI security and risk management sector, providing solutions for enterprises to secure their generative AI applications. The company was founded in 2023 by Ruchir Patwa and Ankita Kumari. Patwa, the CEO, is a cybersecurity expert with over a decade of experience, including leading security teams at Google and Mobile Premier League (MPL). Kumari has a background in product leadership, focusing on fraud and risk management solutions at companies like McKinsey & Company, MPL, and CRED.

The firm's core business involves identifying and mitigating security vulnerabilities in generative AI models and applications. It serves clients across various industries, including e-commerce, travel, and financial services, by helping them preemptively address security blind spots and potential attacks. SydeLabs' business model is consumption-based, charging customers per scan or based on the volume of prompts evaluated. In March 2024, the company secured $2.5 million in a seed funding round led by RTP Global, with participation from Picus Capital and notable angel investors.

SydeLabs developed a product suite designed to offer comprehensive vulnerability assessments for GenAI systems. One of its key products, SydeBox, is used to detect vulnerabilities in AI models before they are deployed into production. Another offering, Syde Guard, acts as a real-time firewall to monitor and protect AI applications post-deployment. The platform provides automated red teaming, which involves simulating attacks to test system defenses, and real-time threat protection. This helps organizations meet evolving compliance requirements from bodies like NIST, MITRE, and OWASP, as well as regulations such as the EU AI Act.

In a significant milestone, SydeLabs was acquired by Protect AI on July 31, 2024. Following the acquisition, the SydeLabs team and its technology were integrated into Protect AI. SydeBox was rebranded as Protect AI Recon and incorporated into the Protect AI platform, enhancing its capabilities in testing and hardening large language models against security risks. This strategic move aims to provide a more comprehensive, end-to-end AI security solution for building, managing, and deploying LLMs securely.

Keywords: AI security, risk management, generative AI, vulnerability assessment, red teaming, compliance, threat protection, large language models, cybersecurity, application security