Vaidas Kondrackis, vice president for risk management at Kneip, a Luxembourg-based data management and reporting solutions provider for the investment and financial industry, spoke with Delano about the firm’s ongoing exploration of generative AI technology, as part of Delano’s series. He highlighted its potential in boosting staff efficiency and enhancing client services. Kondrackis also addressed the various challenges and risks tied to AI implementation, emphasising the importance of responsible and ethical use of AI. Since 2022, Kneip has been a part of Clearstream and Deutsche Börse Group.
Kangkan Halder: Does Kneip utilise artificial intelligence? Is Kneip exploring the use of generative AI?
Vaidas Kondrackis: AI is all about making our people and our systems work better and faster. Kneip has started to explore the potential use of generative AI in the company and client services. We see an opportunity to use AI to help our staff and improve their efficiency. For example, and in very simple terms, as non-native speaker I was using large langue model to review my replies to correct the grammar and make them “sound more English”, a use case that everyone in the company can immediately implement.
Besides use of GenAI, does Kneip employ artificial intelligence for client services?
Kneip is a service provider assisting asset managers and distributors to manage their fund data and supporting regulatory compliance. One of Kneip’s services, offered in partnership with Next Gate Tech, focuses on portfolio data. AI and, more precisely, machine learning algorithms, are used to automate data flows as well as to harmonise and enrich clients’ data.
Are there other specific sectors where you plan to implement AI in the future?
Yes, we are trying to enhance and facilitate the use of our internal policies, procedures and knowledge base using AI. We are planning do so, for instance, by employing retrieval-augmented generation (Rag) in combination with LLM that allows for ‘talk with documents’.
We are also testing how AI can enhance our existing processes and services, with a strong focus on information and data quality, and accuracy.
Organisations must carefully address the associated risks to ensure responsible and effective AI adoption.
What are the main challenges for implementing AI in a company, are there some risks?
Potential implementation of AI in any company presents both benefits and challenges. Every organisation utilising GenAI will need to address AI-associated risks such as:
- Hallucination risk: This occurs when AI provides misleading answers that are factually incorrect but sound highly plausible.
- AI bias and discrimination: These issues can arise from the data used to train AI systems or from inherent biases in the chosen algorithms themselves.
- Security and data privacy risks: AI can generate convincing fake emails by analysing an individual’s online activities. Moreover, AI enables anyone to create realistic fake videos and audios. Additionally, by asking specific questions (prompt engineering), users can extract entire datasets used to train AI, potentially leading to confidential or proprietary data breaches.
- Regulatory risk: The EU’s Artificial Intelligence Act establishes key principles for AI use, including human oversight, i.e., the so called ‘expert-in-the-loop-principle’, transparency, fairness, data privacy and accountability. These principles must be carefully considered when implementing AI solutions.
- Responsible and ethical use: We need to ensure that AI is developed and used in a responsible and ethical way by the company and by the individual employees.
In conclusion, GenAI and machine learning algorithms offer a big potential for enhancing processes and procedures. However, organisations must carefully address the associated risks to ensure responsible and effective AI adoption.
Could you outline Kneip’s short- to medium-term strategy for staying abreast with future AI advancements?
The current speed at which AI is evolving makes it difficult to predict what the next few years will bring us and almost impossible to say how this technology will transform the fund industry in the upcoming decade. One thing however is certain, AI is here to stay and embracing this now--technology as a so called “AI practitioner”--is vital to remain competitive: on the one hand to improve efficiency and customer service and on the other hand, to drive innovation of new products that help asset managers and investors to develop and buy fund products with trust and confidence.