The MNC has partnered a GenAI startup to prove that AI can be implemented reliably, sustainably and securely across its businesses

In the world of legal administration, reviewing and analyzing legal contract provisions requires high levels of accuracy and output quality.

Generative AI (GenAI) has been making inroads in relieving staff from some of the tedium, and one firm is even developing proofs-of-concepts (PoC) for using large language models (LLMs) to produce high quality legal contract reviews with high privacy assurance.

At Jardine Matheson, Group Digital Director Anne O’Riordan is the person helming the PoC project with a GenAI startup in ongoing efforts to develop, test, and deploy AI to position the group’s businesses for the future. “These PoCs tackle real business challenges, enhancing operations by combining high-quality output with resource efficiency, while ensuring the security of our proprietary data. We’re excited to see the results of this engagement and by the potential for future collaboration.”

With the help of the external experts, the firm is fine-tuning the legal output for maximum accuracy and data privacy while minimizing carbon footprints. The project also extends to improving the firm’s customer service levels. The parameters of concern are:

  • The challenges posed by large LLMs — particularly their high costs, data security concerns, and environmental impact
  • Concerns raised by industry studies suggesting that popular LLMs consume 15 times more energy than a typical web search
  • Overcoming the above problems by using smaller and domain-specific LLMs that operate locally within a firm’s own environment (on-premises), to ensure resource-efficient AI and data sovereignty
  • Ensuring that data is not used for training, or retained by the provider of the underlying LLM
  • Leveraging new generation eco-sustainable AI-optimized servers running on processors that offer improved performance per watt across key workloads like AI, data analytics, high-performance computing, and others use cases

According to Prof Lam Kwok Yan, co-founder, TAU Intelligence, the startup firm in the project: “By focusing on domain-specific knowledge, we can fine-tune AI models in order to deploy smaller LLMs in a safe manner, mitigating the problem of hallucination and ensuring the responsible use of AI. The collaboration with Jardines signifies a focus on the development of responsible and resource-efficient AI solutions by companies, setting a benchmark for innovation and sustainability in the AI industry.”