Lakera launches security for generative AI in businesses
19 October 2023 10:51
Lakera has launched an AI-powered threat intelligence software known as Lakera Guard on the market, which the Zurich-based AI security firm founded in 2021 describes in a press release as the “world’s most advanced” solution in this area. It allows developers to immediately protect the AI applications that they create against cyber threats.
The solution is centered upon large language models (LLM), a type of deep learning algorithm that processes text in a human-like manner. These have a significant impact on application areas such as content creation for websites and products, chatbots and virtual assistants, data analysis and customer support.
According to the press release, Lakera Guard is based on a proprietary database that combines crowdsourcing data from Gandalf, the viral IA education game developed by Lakera, with open source databases and the company’s own research. Lakera’s database has amassed nearly 30 million attack data points and is growing by more than 100,000 entries each day. Lakera Guard makes it possible to activate this vulnerability data with a single line of code, instantly embedding a security layer into AI applications in production.
“AI presents security risks unlike anything we’ve seen before”, comments Harald Nieder of Redapline, the lead investor in Lakera. He adds that Lakera boasts “one of the most experienced teams in AI with a treasure trove of vulnerability data”, while the company’s “developer-first approach provides the missing link for all enterprises to deploy AI models with confidence”. ce/mm