The first day of this year’s unique ICT Spring edition brought together local and international experts around the topic of Artificial Intelligence. The morning conference entitled “AI, the game changer”, which took place on September 15th, had experts discuss the latest digital innovations, with a focus the post-Covid-19 crisis and on the latest AI applications.
“Times are challenging, but we have proven that we can adapt: more connected, more conscious, more digital, but also more Human. Be resilient, but also antifragile”: this is how Kamel Amroune, CEO of Farvest Group and co-founder of ICT Spring, welcomed the participants of the 11th edition of the international tech summit. Charlotte Boutelier, Head of Marketing, Farvest Group, then explained the new phygital concept to all the attendees. The organizers then welcomed Jonathan Prince (Co-founder, Finologee) on the stage. He acted as Master of Ceremony for the morning session and started by asking several questions around the topic of AI: “is the hype around AI fully justified? Is there such thing as an AI? Can only big companies afford it? How is it transforming the way we work and does it change our daily lives?”
Michael Mossal (Member of Innovation & Technology Community, CTO, NTT ltd. Europe) then took the stage to give a keynote speech entitled “Digital Innovation - How Luxembourg exports digital by staying innovative!”. He first described innovation in the words of one of his peers: “having an idea and being bold enough to get the idea through”. According to the expert, companies need to invest into the new world, or at least, try to link their world to the new world. Others companies have evolved, going from one tech to another, moving from one decade to another. Michael Mossal than presented some of the innovative projects NTT has been working on in Luxembourg: “the government of Luxembourg issued a call for projects around the topic of 5G: we defined an IoT platform for automotive and sports, and are also providing an innovative tech to the government of the Isle of Man to make the TT race safer”. He then presented a couple of health projects developed in Luxembourg: the large testing scale in the frame of the Covid-19 pandemic – to figure out where the virus can spread – and the creation of a medical product catalogue.
“Cognitive Automation – bots liberating employees and reimagining jobs” was the topic addressed by Vitalie Schiopu (AI Architect and Product Manager, Accenture). He first defined “cognitive automation” as a virtual workforce of bots, which can run part of your business organizations, before sharing the results of a large survey conducted by Accenture: 84% of executives say they won’t achieve their growth objectives without scaling AI, and 76% say they know how to plot…but struggle to scale AI across the business. “Cognitive automation allows machines to mimic what Humans can do: interpret context, monitor processes; dialog in natural language. “Tapping it into the workforce will complement and emulate the Human workforce and will allow employees to do things differently, be more efficient and even create new products and services”, added Vitalie Schiopu. He continued: “it is the winning formulae of people and technology to enable key transformations: automation, augmentation and innovation”. When it comes to innovation, he explained that it can actually fee up existing people’s time to work on new priorities and give them the ability to pivot to new areas.
Candi Carrera (Country Manager, Microsoft Luxembourg) then focused on “the AI benefits and lessons from the Covid-19 crisis”. He first stated: “this impact had a huge economic impact with the global trade seeing a huge decrease and several industries being hit. Also, 80% of the global workforce had to work from home and offices had to shut down. And there are also important social impacts due to the lockdowns, mobility restrictions, school shutting down, etc.” The expert then shared several examples of AI supporting scientific and medical activities (chest scans, chatbots for emergency services, etc.), big data being used by governments, and the fact that there has been a huge teleworking and digital acceleration. The Country Director then shared several examples: school in Luxembourg had 99,000 daily Teams users and AI blurring the background, the combination of big data and AI to support political decisions makers (with the example of wastewater plants to monitor the evolution of the virus) and healthcare bot service with the implementation of virtual agents allowing for self-assessment. “In the post-Covid period, behaviors will change, with more phygital events, vertical distance, people working from home, a knowledge based on data, and with the advent of robots to support Humans. Moreover, we have seen many benefits and positive uses of AI, with valid concerns. We therefore need to continue to monitor and be cautious about how we use this tech”, concluded Candi Carrera.
Liv Boeree (Speaker, Astrophysicist, Philanthropist & Poker Champion) then explored the evolution of AI through Gaming. The professional poker player first explained that games can be defined as problems to be solved, with specific goals to achieve. “It is a good way to master decision-making skills” she underlined. According to her, poker is a game of hidden information, but also a game of luck and skills: intuition-based skills are needed and so are analytical skills. “The game used to be very instinctive, but has now switched to analytics and mathematics. Players can now understand the game on a much deeper level. This is where AI comes in as it is able to calculate and implement maths solutions in real time” added Liv Boeree, who shared multiple examples of machines beating Humans in different games, from poker to Go. What does it mean for professional gamers? “I expect poker to move away from the online sphere and become more old school, with people playing face-to-face, trying to read each other’s game”.
“Managing the AI process: putting humans (back) in the loop” was the name of the presentation given by Dr. Balázs Kégl (Head of AI research, Huawei France), whose main objective as the head of Huawei’s Noah’s Ark in Paris is to optimize engineering systems by making them better, cheaper, more reliable, safer and more energy efficient. “There have been highly visible recent AI breakthroughs, but why aren’t these advances already in engineering systems?” asked the expert. He then described a typical engineering control system, giving notably the example of an airplane, with the engineer controlling the system, observing performance indicators and tuning some parameters from time to time. He added: “automated control, if exists, is based on a deep understanding of the physics of the system. In this context, AI learns the system behavior based on historical data and uses it for better control”. After describing the roles of the data scientist and of the system engineer, he explained that they need to collaborate and trust each other. Dr. Balázs Kégl ended his presentation by sharing his recommendations: a strong top-down mandate, form “commandos”, build trust through iterative pilots, design and enforce the data science process and finally start building standard operational tools for data science.
Massimo Moggi (President & CEO, WESTPOLE, Professional Affiliate Scuola Superiore Sant’Anna di Pisa) recently published a research paper on the economics of AI and shared some of its results. “AI has a long history but recently became a key tech driver. It is due to the explosion of data, computer power, etc. There is a correlation between AI and the affirmation of the internet society”, said the expert. He described AI as a niche market with many applications: computer vision, speech processing, NLP, etc, with China and the US dominating the market. Massimo Moggi also focused on the economic impact of AI on the labor market: “studies are expecting a 10-15% job loss in the next 10 years, mainly impact call centers and production chains. In Europe, we are also expecting an important gap in the supply of new profiles. What should be done at schools? We need to work with the governments to requalify and restructure job profiles. A deeper collaboration is needed between the public and private sectors”.
How Luxembourg is building a smart nation and a data-driven economy
In a video, Carlo Thelen (Director General, Luxembourg Chamber of Commerce) first stated that he was glad to see that ICT Spring was able to transform into a premium global digital event: “It shows the importance of digital in today’s world. But it still has its limitations as some face-to-face interactions with Humans cannot be replicated. I hope we can all return to a more Human way of working and exchanging soon”. Carlo Thelen then highlighted that companies were able to adapt notably by using collaborative digital tools, but that the risks linked to homeworking are still unclear. “There has also been an accelerated digitalization which allowed companies to ensure business continuity. Therefore, digitalization and simplification will be two conditions for socio-economic development and will support the post-Covid-19 recovery phase,” he added. According to him, it is key to reduce the administrative burden by simplifying procedures. “Luxembourg is on the path to being an advanced digital society as it builds a sustainable data-driven economy and citizen-focused AI. Also, the country has to lean on several pillars – innovation generation, talent competitiveness, fiscal attractiveness, and more. It needs to focus on developing new sectors and diversifying its economy. ICT Spring is instrumental in creating and generating new ideas and innovations,” concluded Carlo Thelen.
The organizers then had the privilege to share a video of Xavier Bettel, Prime Minister of Luxembourg, who has been a great supporter of ICT Spring for the past few years. “We live in a strange time. Organizing ICT Spring like it is usually done would have been impossible this year. Without digital, it wouldn’t have been possible: it’s also the case with school, medicine, government meetings, etc. The digital world is a reality and we know that Luxembourg is a smart nation!” added the Prime Minister.
AI at the Center of Digitalization - Managing robots, algorithms, AI… and people
The morning session ended with Peter Sondergaard (Founder & Executive Advisor, The Sondergaard Group). “AI Adoption has been increasing the last four years both in terms of number of organizations using AI, number of models deployed by organizations and the breadth of organizational departments using AI models” started the former Gartner VP. According to him, the first phase of digital transformation ended in 2020 and due to the Covid-19 pandemic we saw and acceleration of digital, but not a full implementation of what the entire strategy was. It was just a move to agile strategy. He asked: “but what will happen in the next phases when the cloud is pervasive? When AI is pervasive? When more technologies will emerge? Imagine this world”. How to run such organizations? AI leadership is needed. “Strong leadership teams and strong middle management teams are the differentiators. Classic leadership skills need to evolve. How can we be agile in mindset and execute in a non-traditional way?” added Peter Sondergaard who continued: “first, the CEO or senior leader in a government institution needs to be the person ultimately accountable for the AI Governance charter as well as a clear assignment of responsibility within the organization. Other functional executive roles could have specific responsibilities around AI governance, but this is where AI, including AI governance, becomes a general leadership skill that every manager in an organization needs to master”. He then shared the seven new leadership skills: technology, data science, agility and AI – the hard skills – and storytelling, ideation and innovation as soft skills. When discussing AI governance, which he defined as ensuring that AI is explainable, transparent, accountable and ethical – he shared the keys to developing and implementing an AI governance charter. This charter is composed of six key elements: right to know, notification of AI decisions, review of AI decisions, opt-out of AI systems, notification of bias and remedy for misuse. Finally, Peter Sondergaard explained that AI leadership and governance need to be measured: “define what the world will be like with AI in the center of what you do”.
Photos: Dominique Gaul
Publié le 16 septembre 2020