One of the primary value propositions of AI platforms is their capacity to summarize, evaluate, and answer questions depending on user input. The service's ability to effortlessly ingest and handle a variety of file types is crucial. However, if files are not adequately inspected, attackers can take advantage of this feature by inserting harmful scripts in seemingly innocent files such as images or PDFs. Distinct from prompt injection, this is sometimes called “indirect prompt injection.” It was critical for this AI company to have a comprehensive file analysis solution with:
- Superior malware detection to stay ahead of evolving cyber risks
- High-performance scalability to handle platform usage which was growing at an exponential rate
- Easy deployment and integration which require minimal administrative overhead
High-Fidelity Intelligence and Accurate Detection
Spectra Intelligence leverages the most up-to-date global file and network reputational data to classify files as goodware or malware. It assesses documents, images, and other files that users upload to the AI platform in real time. Files classified as malicious are flagged within the organization’s security workflow for appropriate handling. With millions of new samples and network IOCs added daily fuelling RL’s industry-leading Threat Repository, Spectra Intelligence can identify hidden threats that may go undetected by other security solutions.
Effortless Scalability
RL’s easy to use APIs enabled the company to integrate Spectra Intelligence with their production environment and workflows without major changes. The high-performance, high-volume query processing delivers results in milliseconds and is optimized to handle large spikes in file upload volumes. Monthly usage has grown rapidly to over 1.7 billion API calls with minimal administrative overhead, demonstrating the scalability of Spectra Intelligence.
