HoundDog.ai raises $3.1M to help companies proactively prevent sensitive data leaks – Business
HoundDog.ai Inc., a startup that has built an artificial intelligence-powered code scanner to prevent sensitive data leaks and automate compliance, said today it has closed on a $3.1 million seed funding round.
Today’s round was led by E14, Mozilla and Ex/ante and saw the participation of several angel investors, including Amjad Masad, chief executive of the developer tooling startup Replit Inc. Alongside the funding, the company announced the first version of its platform, generally available now.
The HoundDog.ai cloud platform is designed to help companies apply data security and privacy controls at the time their developers are writing the code for their software and applications. It explains that this is a more proactive approach to security that not only reduces the costs associated with fixing vulnerabilities, but also enables organizations to adopt a “shift-left” strategy for protecting sensitive information and ensure they remain in compliance with regulations.
The startup said it’s necessary to focus on personally identifiable information because this accounted for 92% of all data compromised in security incidents in 2023. Citing a report by IBM Corp., it said these kinds of leaks have a negative impact on customer relationships and can be extremely expensive to remediate.
HoundDog.ai’s code scanner works by continuously searching for vulnerabilities as developers are writing their code. It focuses on exploits that will expose sensitive data as plaintext across logs, files, tokens and cookies, and through third-party systems.
In addition, the platform can track and visualize sensitive data as it flows through an organization’s information technology systems in real time. It can scan more than 3 million lines of code in under three minutes, documenting data processing activity at rapid speed, the company said. Whenever it comes across new data elements, it will categorize them based on their level of sensitivity and automatically create a Record of Processing Activity report for compliance purposes.
AI algorithms are used to magnify the platform’s capabilities and improve the accuracy of its detections. In particular, HoundDog.ai said it uses generative AI to evaluate tokens for sensitive data handling with higher precision rates.
HoundDog.ai founder and CEO Amjad Afanah said there’s an urgent need for companies to become proactive in their efforts to identify and fix code vulnerabilities. According to him, most sensitive data leaks that occur via logs, files and third-party systems are only identified after the fact, meaning the damage has already been done.
“Additionally, documenting data flows for GDPR compliance proves to be a manual, error-prone process that often fails to accurately reflect the reality of constantly changing codebases,” Afanah said. “With HoundDog.ai, companies can take a proactive stance in ensuring their sensitive data is protected and eliminate costly, reactive and error-prone processes for privacy compliance.”
With its launch into general availability today, the HoundDog.ai platform can send its consolidated view of vulnerabilities across all code repositories to users via Jira and Slack, alerting them to any actionable security items. Previously, the platform surfaced its security findings only within user’s continuous integration pipelines, catering to companies that have standardized their security workflows in GitHub and GitLab. By sending its notifications to Jira and Slack, HoundDog.ai said, it can better cater to customers that rely on ticketing systems to report vulnerabilities.
Replit CEO Masad said the startup caught his attention due to the rapid adoption of generative AI coding assistants such as GitHub Copilot. “As an increasing number of companies turn to AI-generated code to accelerate development, embedding security best practices and ensuring the security of this code becomes essential,” he said.
HoundDog.ai’s platform has been given the thumbs-up by a number of early adopted, chief among them Juvare LLC, which develops software for emergency management, preparedness and response.
Juvare Chief Information and Security Officer Bryan Kaplan said it has become essential for companies to take the initiative in terms of code security. But to do this, he said, they need a comprehensive scanning tool that works in real-time and integrates with existing security platforms.
“It needs to provide peace of mind by ensuring that sensitive data does not accidentally leak into logs, files or other systems, even with high-frequency updates to the codebases,” Kaplan said. “It’s these capabilities that drew us to HoundDog.ai, as no one else is presently offering this.”