Kong has introduced updates to its AI Gateway, a platform for governance and safety of LLMs and different AI assets.
One of many new options in AI Gateway 3.10 is a RAG Injector to cut back LLM hallucinations by routinely querying the vector database and inserting related information to make sure the LLM is augmenting the outcomes with recognized data sources, the corporate defined.
This improves safety as effectively by placing the vector database behind the Kong AI Gateway, and likewise improves developer productiveness by permitting them to deal with issues aside from making an attempt to cut back hallucinations.
One other replace in AI Gateway 3.10 is an automated personally identifiable data (PII) sanitization plugin to guard over 20 classes of PII throughout 12 totally different languages. It really works with most main AI suppliers, and might run on the international platform degree in order that builders don’t have to manually code the sanitization into each software they construct.
In keeping with Kong, different comparable sanitization choices are sometimes restricted to changing delicate information with a token or eradicating it completely, however this plugin optionally reinserts the sanitized information into the response earlier than it reaches the top person, making certain they can get the info they want with out compromising privateness.
“As synthetic intelligence continues to evolve, organizations should undertake sturdy AI infrastructure to harness its full potential,” stated Marco Palladino, CTO and co-founder of Kong. “With this newest model of AI Gateway, we’re equipping our prospects with the instruments essential to implement Agentic AI securely and successfully, making certain seamless integration with out compromising person expertise. Furthermore, we’re serving to resolve a few of the largest challenges with LLMs, resembling reducing down on hallucinations and bettering information safety and governance.”
