Interview: Causality Link on what testing genAI says about future deployment

Read the full article on the Finadium website (subscription required).

Since its launch last year, OpenAI’s ChatGPT bot has sparked huge interest across financial services in the potential of large language models (LLMs) and generative artificial intelligence (genAI), albeit with cautious conservativism in terms of implementation. We speak with Pierre Haren, co-founder and CEO of explainable AI research platform Causality Link, about the way these technologies are being used after the team launched a free trial and what could ignite developments towards SecFinGPT.

Causality Link’s integration of LLMs into its research platform has started small, with its signal-detection mechanism being augmented to produce digestible, precise text about companies. In other words, the underlying algorithm detects novelty in trends, while the LLM model creates a brief analytical summary that combines multiple, curated perspectives, which are then distributed to users, explained Haren.

The team’s methodology and results are published in a recent paper, Analytics Controlled Narratives with Structured Retrieval Augmented Generation (SRAG), showing how this approach ensures “accuracy and contextual relevance” by merging “the best of generative AI with symbolic AI”.


Recent Posts

Follow Us