DigitalTransformationTalk: WEF Special Episode - Navigating AI at scale - towards successful implementation, optimisation gains and frugal digital transformation

On 16 January 2024, Digital Transformation host Geoff White was joined by: Jose Saiz de Omeñaca Monzón, Expert - Procurement / Supply Chain / Sustainability / Ecommerce /AI United Nations Economic Commission for Europe, Freelance; Martin Miller, Ex-Director of AI/ML Production Operations, Levi Strauss & Co; and Dr. Peter Weckesser, Chief Digital Officer, Schneider Electric

Views on news
As on a panel in Davos Julie Sweet of Accenture explained, there are thousands of use cases for artificial intelligence in business. However, before implementing AI, it’s key for C-suites to have a deep understanding of the technology. With previous new technologies as the PC or the cloud, the responsible use of the technology wasn’t a central issue, while it’s of the utmost importance in the case of AI. There is also a need for reframing how businesses and governments think of talent and retraining to smoothe out disruptions that automation is about to cause. Some of the companies have already realised that the management of AI must come to the forefront and have appointed their chief-AI-officers. Schneiders chief-AI-officer is in contact with everyone in the company on every level.  They use other companies foundation models and finetune them to meet their use cases. Analogies can go a long way when IT experts are trying to explain AI. Currently, we are in a stage of goldrush, where some businesses will find gold, others just pyrite. To ensure that the business is on the right path, you need metrics and measure the performance of the AI technology against them and readjust when necessary. 

How to find the right talent for AI deployments
AI is already implemented by several big fashion brands. It’s also important for leaders to bear in mind that an AI model won’t operate well until it has been trained to a specific use case. Businesses may need expert recruiters to hire workforce with advanced digital skills., who aren’t easy to find either. Equally important are good analytic and decision making skills with data. But you also need tools and solutions that come between data scientist and productionisation. For building ML models internally, you will need top level experts. Even if you decide to use an open source large language model and customise it for a chatbot tool, you need to ensure you have a good quality data flow and that you don’t increase your reputational or security risk with new GenAI tools. Non-tech companies don’t have to reinvent the wheel, they can procure AI products off-the-shelf – many of themsupplied  by cloud service providers. 

The panel’s advice

• Find dedicated recruiters who can bring the right digital talent on board.
• Before recruiting digital talent, assess your skills needs in the context of your business strategy, factoring in attrition and retirement rates too.
• For training purposes, companies with more resources can set up learning platforms and leverage gamification.
• Software developers are more open to new technologies, but you must make it clear that you aren’t after new technologies to be able to replace them but to achieve productivity gains.