Galileo Releases Evaluation Foundation Models for GenAI
Generative artificial intelligence (GenAI) developer Galileo has released a suite of evaluation foundation models (EFMs) designed to help enterprises bring trustworthy AI into production.
The new Galileo Luna EFMs are designed to make GenAI evaluations faster, more cost-effective and more accurate, the company said in a Thursday (June 6) press release.
“For GenAI to reach mass adoption, it’s crucial that enterprises can evaluate hundreds of thousands of AI responses for hallucinations, toxicity, security risk and more, in real time,” Vikram Chatterji, co-founder and CEO of Galileo, said in the release. “In speaking with customers, we found that existing approaches, such as human evaluation or LLM-based evaluation, were too expensive and slow, so we set out to solve that.”
The Galileo Luna models are purpose-built for GenAI evaluation, and each one has been designed for specific evaluation tasks, according to the release. Because they are smaller and purpose-built, these models provide greater accuracy, speed and cost-efficiency.
Luna has been integrated into all Galileo platforms to help these tools intercept harmful inputs, improve system security and enhance operational efficiency, the release said.
These EFMs are already being used by Fortune 50 consumer packaged goods (CPG) brands and Fortune 10 banks to handle millions of GenAI application queries per month, according to the release.
Alex Klug, head of product, data science and AI at HP, said in the release: “Until now, existing evaluation methods, such as human evaluations or using LLMs as a judge, have been very costly and slow. With Luna, Galileo is overcoming enterprise teams’ biggest evaluation hurdles — cost, latency and accuracy.”
In another recent product launch, Galileo said in February that it was releasing a new retrieval augmented generation (RAG) and agent analytics solution designed to help businesses develop trustworthy AI solutions.
PYMNTS Intelligence has found that while GenAI is revolutionizing finance and banking by enhancing customer interactions and risk models, the technology also introduces challenges like data security and systemic decision-making risks.
As a result, regulators and others are taking actions to address risks associated with GenAI, particularly in areas of model explainability, according to “Banking on AI: Financial Services Sector Harnesses Generative AI for Security and Service,” a PYMNTS Intelligence and AI-ID collaboration.
