In a significant development in the cybersecurity landscape, Repello AI—a deeptech startup focused on securing generative AI systems—has raised $1.2 million in a seed funding round. The round was led by Antler, with participation from strategic angels across the U.S., Europe, and India. The capital will be used to further develop Repello’s automated red teaming platform, aimed at proactively identifying and mitigating vulnerabilities in GenAI deployments.
Tackling the Hidden Threats of Generative AI
As businesses increasingly adopt generative AI tools across functions like customer service, content creation, and automation, the risks associated with these systems are also growing. From jailbreak prompts to hallucinated outputs and data leakage, GenAI models are susceptible to a new class of threats that traditional cybersecurity measures struggle to address.
Repello AI steps into this gap with an automated “red teaming” solution. Traditionally a human-led process, red teaming involves ethically simulating attacks on AI systems to uncover weaknesses. Repello is scaling this process by building an automated, continuous testing framework to act as a guardrail for GenAI tools in real-time environments.
According to Repello, the platform can identify both known and novel failure modes in large language models (LLMs) and other GenAI applications, helping enterprises remain compliant and resilient as regulations around AI safety tighten globally.
The Core Team and Vision
Founded in 2023 by Aditya Vempaty (former CMO at Observe.AI and ex-Google), Aditya Pamaraju (ex-AWS and Palantir), and Ayush Sood (ex-Meta), Repello AI brings together expertise in cloud security, AI systems, and enterprise SaaS.
“Our goal is to build a scalable safety layer for generative AI models being used in production environments. Today’s enterprise-grade systems require not only innovation but also accountability and robustness,” said Vempaty in a press statement.
The team operates out of San Francisco and Bengaluru, indicating a dual-market focus that aligns with both global and Indian enterprise demand for secure AI infrastructure.
Growing Market for AI Guardrails
The funding comes at a time when AI safety is emerging as a global priority. With regulatory frameworks like the EU AI Act and increasing scrutiny from data privacy bodies, companies are under pressure to ensure their AI deployments are not only effective but also secure and ethical.
Recent reports suggest that enterprises using GenAI at scale have witnessed a sharp increase in hallucination-related risks, where AI generates plausible but inaccurate content. These can lead to misinformation, reputational damage, and even legal challenges. Repello’s platform aims to mitigate these risks by simulating adversarial attacks and flagging system weaknesses before they reach production.
Industry experts note that red teaming will become a non-negotiable aspect of responsible AI deployment, especially in sectors like banking, healthcare, and government services where data sensitivity is high.
What the Investors Are Saying
Antler, the lead investor, emphasized the timeliness and importance of Repello’s mission. “As the adoption of GenAI accelerates, there is a parallel need to build strong safety infrastructure. Repello is building what every CTO will soon require—automated, scalable, and continuous security testing for AI systems,” said an Antler spokesperson.
Other strategic investors include former senior executives from Salesforce, Microsoft, and OpenAI, indicating confidence in Repello’s market relevance and technical depth.
Looking Ahead
With this funding, Repello plans to expand its engineering team, enhance the capabilities of its testing framework, and onboard pilot enterprise clients. The platform will also aim to support more GenAI models and integrate with popular AI development stacks.
The startup is expected to formally launch its enterprise offering later this year, positioning itself as a core part of the GenAI safety ecosystem.