Prompt Security has emerged from stealth mode with a new solution powered by artificial intelligence (AI) to protect companies’ AI products from prompt injection and jailbreaks, as well as prevent employees from mistakenly exposing sensitive data to tools like ChatGPT.
While organizations are eager to reap the benefits of generative AI (GenAI), they are increasingly concerned about the potential impact of these cutting-edge tools on their security infrastructure. A recent Dark Reading survey revealed that respondents identified several risks associated with fully embracing GenAI, including the opacity of third-party tools, a lack of consensus on GenAI guidelines and policies, and data governance concerns.
Prompt Security aims to address these risks by implementing measures to safeguard every interaction with GenAI within organizations, from internal AI tools to commercial products with AI features. The company’s solution meticulously checks each prompt and model response to prevent exposure of sensitive data, block harmful content, and thwart GenAI-specific attacks. By analyzing semantic data, Prompt Security can mitigate threats such as prompt injection, jailbreaking, and data extraction. Additionally, contextual large language models (LLMs) are utilized to detect and redact sensitive data, safeguarding customer and employee information, as well as intellectual property, from unintended exposure.
Furthermore, the solution provides comprehensive documentation of the variety of AI tools utilized throughout the organization, allowing the security team to monitor how these tools are being used and establish access policies based on the application and user group.
Prompt Security’s co-founders, who previously worked at Orca Security, have significant expertise in the field of AI and security. CEO Itamar Golan was previously the head of ML and AI at Orca Security, while CTO Lior Drihem served as the head of innovation. The company recently secured a $5 million seed round led by Hetz Ventures, with participation from Four Rivers and multiple angel investors.
With the launch of Prompt Security, organizations grappling with the potential security implications of integrating GenAI into their infrastructure can leverage AI-powered protection to minimize risks and ensure the integrity of their AI products and sensitive data. The company’s innovative solution offers a proactive approach to addressing the security challenges posed by the rapid evolution of AI technology.