Here’s the translation into American English:
—
In an environment where generative artificial intelligence is revolutionizing various industries, organizations are increasingly interested in leveraging its potential. However, the transition from ready-to-use solutions to large-scale implementation brings operational and technical challenges that need to be addressed. This analysis focuses on key learnings from AWS customers in Europe, the Middle East, and Africa who have successfully navigated these issues, providing a framework for those looking to follow this path.
The foundation of successful generative AI implementations lies in creating business cases with clear value propositions that align with organizational goals, such as improving efficiency, reducing costs, or increasing revenue. Examples include enhancing customer experience and optimizing operations.
In EMEA, various companies have used AWS services to transform their operations. For instance, Il Sole 24 Ore, a prominent media group in Italy, partnered with AWS to enhance an outdated service that allows users to make tax inquiries and receive answers from experts. Through a Retrieval-Augmented Generation (RAG) solution, they achieved 90% accuracy in responses, reducing search time so experts could focus on strategic tasks.
Booking.com, a global leader in travel services, has utilized scalable generative AI technology to create personalized experiences for its customers through Amazon SageMaker AI. According to Rob Francis, their Chief Technology Officer, he values the flexibility AWS offers and the importance of open-source software in advancing generative AI.
ENGIE, a global energy company, developed a chatbot that enables conversational search within its data hub, making it easier for users to locate information across thousands of datasets. This tool has accelerated data-driven product development and improved sharing within the organization.
While a solid business case is crucial, transitioning to generative AI poses new challenges, such as scalability and data governance. Adopting a holistic approach that considers not only technological aspects but also effective production infrastructure is vital.
The importance of establishing quality standards in solution development is evident in cases like that of the Iveco Group, which implemented a cloud operating model that optimizes its developers’ time. Another example is the Accor Group, which applied fundamental software development principles when designing a generative AI-driven reservation application, ensuring service quality through a thorough testing process.
Danske Bank benefited from a modular architecture with AWS, enabling it to integrate various generative AI tools and services. Meanwhile, the Schaeffler Group developed a comprehensive framework that establishes security measures for large-scale implementation.
With generative AI applications handling sensitive data, security must be a priority. This involves establishing access controls, data encryption, and access monitoring. Companies like Il Sole 24 Ore are implementing self-regulation codes to ensure ethical use of AI, while Accor has set measures to keep its chatbot functioning within ethical boundaries.
The transition from pre-production to large-scale deployment of generative AI applications presents both challenges and opportunities. Identifying a strong business case and maintaining high infrastructure standards are critical. Companies in EMEA have demonstrated that by utilizing AWS services, it is feasible to overcome obstacles and maximize the benefits of generative AI in a responsible and efficient manner. This allows more organizations to benefit from this transformative technology.
via: MiMub in Spanish