The unprecedented growth of GenAI holds the potential to disrupt nearly every industry - promising both competitive advantage and a creative edge for its users. With over hundreds of AI and Machine Learning (ML) tools already existing in the market, a number that continues to grow, the opportunities are almost limitless.
Innovative AI startups are emerging to solve industry problems, while new sets of Large Language Models (LLMs) are being introduced every week with new capabilities.
To leverage the increasingly fast paced evolution of these technologies, enterprises need to invest in core capabilities and maintain an adaptive, agile and experimental growth strategy. We expect that by 2025, the advancement of GenAI will allow enterprises to run advanced industry-specific use cases, with the right level of accuracy in a cost-effective manner, ultimately automating existing processes.
One of the main areas of consideration for enterprises is the differentiation between commercial versus open source LLMs that can be adopted as core technologies in their ecosystems.
Commercial LLMs are proprietary and they are available via service providers and are usually on consumption basis. They are being hosted by the service provider or in the service provider’s environment, may limit the potential usage in certain cases
Open source models, on the other hand, can be hosted in a customary environment and can be the right choice when the privacy and ownership of data is in a requirement. When it comes to open source, there are a number of points to be considered.
The licensing restrictions around the commercial use of open source models and the organisation who provides the model needs to be credible to ensure the quality of data behind the model.
Some of the best open source models are Llama 2 by Meta or Falcon by the Technology Innovation Institute, based in the UAE. Falcon 180B is a powerful language model with 180 billion parameters, trained on 3.5 trillion tokens and is currently one of the best open source models and can be used for both research and commercial use.
These models can be fine tuned to be used for code generation, content creation or translation. In the creative domain, we have models for creating images, video generation or even music and now we’re starting to see some new industry specific models, especially in the healthcare, finance and legal domain.
Enterprises need to develop a both near-term and long-term strategy to embark on this transformation, based on their readiness, technology and data capabilities.
So, what are the pathways to GenAI adoption?
AI Augmentation (Short term): By adopting current AI tools, employees can access real-time insights, knowledge, and support systems that allow them to be more productive. On the customer service front, AI can offer personalised interactions, quicker response times, and more accurate solutions, ensuring a heightened user experience.
AI Transformation (Medium term): Automate and elevate existing complex business processes by reducing manual steps and human dependencies, enhance data and document processing. By connecting existing processes to GenAI platforms, businesses can achieve improved efficiency and productivity while reducing human errors.
AI at Core (Long term): Instead of treating AI as an add-on, businesses will embed AI functionalities at the core of their platforms and operations. This deep integration ensures that AI is a foundational element of the organisation.
The impact to the workforce and to business and operating models in an enterprise
The adoption of AI will inevitably transform the workforce landscape. While some tasks will be automated, new roles and skills will be created and emerged. The introduction of AI tools necessitates upskilling and reskilling. Employees must be equipped not only with the knowledge to work alongside AI but also an understanding of its risks and limitations.
Conclusion
GenAI is one of the biggest paradigm shifts of our generation. It will introduce a significant opportunity for empowerment and transformation. There is the need for responsible AI from the risk and control perspective to address potential privacy, safety and security issues. GenAI can equip people with not just information, but also specialised skills accessible to anyone and help maximise human potential.
