Artificial intelligence and machine learning cybersecurity company Protect AI Inc. today announced that it has acquired artificial intelligence security startup Laiyer AI for an undisclosed sum.

Layier AI’s core product, LLM Guard, is designed to provide security for large language models such as OpenAI LP’s GPT-4 for various security threats, misuse and attacks. LLM Guard protects against threats such as prompt injection attacks, data poisoning and other vulnerabilities that can arise in AI models.

Prompt injection attacks involve attackers manipulating LLMs through crafted inputs, leading to data exposure or decision manipulation. The attacks can be direct, via the LLM’s input, or indirect, through tainted data sources. They often bypass detection because of the implicit trust in LLM outputs.

LLM Guard can detect, redact and sanitize inputs and outputs from LLMs. Doing so mitigates risks that can manipulate a model’s outputs to personal data leaks, ensuring the integrity and safety of the data processed by these models. The service is optimized for performance and integrates with existing security workflows, offering tools for observability, such as logging and metrics.

According to Laiyer, LLM Guard provides a threefold reduction in CPU inference latency, enabling the use of cost-effective CPU instances instead of expensive GPUs without compromising accuracy. The code is open-source and has decent adoption, with Laiyer AI seeing more than 13,000 library downloads and 2.5 million downloads of its models on Hugging Face in just 30 days.

With the acquisition, Protect AI intends to create a commercial version of LLM Guard with expanded features, capabilities and integrations within the Protect AI platform. “These new capabilities will empower our customers in automotive, energy, manufacturing, life sciences, financial services and government sectors to develop safe, secure GenAI applications,” Chief Executive Ian Swanson said ahead of the announcement.

Laiyer co-founders Neal Swaelens and Oleksandr Yaremchuk added that “by joining forces with Protect AI, we are extending Protect AI’s products with LLM security capabilities to deliver the industry’s most comprehensive end-to-end AI security platform.”

Protect AI is a venture capital-backed startup, having last raised $35 million in funding in July. Investors in the company include Evolution Equity Partners LLP, Salesforce Ventures LLC, Acrew Capital LP, Boldstart Ventures LLC, Knollwood Capital LLC and Pelion Ventures Partners LLC.

Image: Layier AI