Artificial Intelligence revolution… or just a dream?

In a previous article, Patrick Darmon, a partner at Fizz venture, asked how to integrate artificial intelligence into his business model. In this second part, he wonders if artificial intelligence could be just a dream.

Before we delve into each of the issues we identified in the article published on March 22, Can artificial intelligence be used in business? , it is important to address the motivation for such an exercise, namely: should companies really invest in AI and use it as an important driver for innovation and transformation? Instead, are we at the end of a bubble with no future in which we have already exhausted the main use cases for AI? A converging body of case studies and empirical findings supports an optimistic future scenario for artificial intelligence:

First of all, luckily there are companies that have been investing heavily in AI and reaping significant benefits from it for years. These are obviously the GAFAMs, even if, as mentioned earlier, they are not comparable in some areas to our large industrial and service groups (especially in terms of market, technology culture, data, regulations, etc.) that have been able to leverage AI to scale their business models, allowing their core business to accelerate growth: product recommendations for, search and advertising for Google, and customer insights and advertising for Facebook. Their AIs have been in production for years and are the originator of the concept of “scaling”.

Then the automotive sector which is experiencing the twin revolution of electricity and artificial intelligence with Tesla (one could argue that Elon Musk is the new Henri Ford, but it is too early to be sure even if Tesla it's already a respectable company) as well as Chinese players such as BYD which is now, with increasing success, penetrating the auto industry's birthplace and disrupting an industry noted for its complexity and capital intensity. When it comes to the use of artificial intelligence, it is probably one of the most successful industries in terms of its ability to integrate it into a complex, high-risk environment.

Artificial intelligence is already making a significant contribution to scientific progress and by extension to R&D functions, as it already does in molecular biology. Until then, all studies converged to indicate that scientific productivity – measured by the number of patents filed per researcher – had stagnated for several years in many industries where the increase in the number of patents came essentially from the increase in the number of researchers! If this indicator seems far from our daily concerns, the fact remains that it is, according to economists, one of the most important indicators for predicting our economic growth. Today, artificial intelligence is changing that! Alphafold, DeepMind's application that accelerates the discovery of protein folding structure can be considered one of the biggest revolutions in artificial intelligence. The coming wave the recent book by Mustapha Suleyman (co-founder of DeepMind) and Michael Bhaskar describes the prospects for convergence between AI and Biology. In the field of new material discovery, artificial intelligence seems to be opening a path similar to that of Alphafold in molecular biology, DeepMind's recent breakthrough in this field already suggests multiple innovations and opens up prospects for other industries.

In several areas, there is artificial intelligence that brings value, such as supply chain optimization for FMCG products, the use of “digital twins” for management and operation in industries with physical infrastructure, the management of insurance claims or the reduction of damage in food distribution. These use cases go beyond the scope of AI targeting limited activities and carry transformational potential that has yet to be tapped. Several large companies are already lighting the way for others: Ping An, mentioned above, which has developed numerous uses of artificial intelligence in risk management, health, and even detecting greenwashing; We can also note Walmart building artificial intelligence tools to optimize its supply chain or even reduce “breakage” (losses related to spoilage of fresh products).

It is still too early to realistically assess the impact of genetic AI on business – even if the exercise is tempting and has already been attempted – however, it seems clear that it will bring productivity gains at work, providing one or more assistants for every . We can also foresee a particularly strong impact of ChatGPT and other LLMs on many service activities.

All these elements are converging towards accelerating the spread of Artificial Intelligence in the company. And this without anticipating the emergence of new technologies resulting from current knowledge in the field. The recent report of the Commission on Artificial Intelligence also moves in this direction and proposes measures to achieve this acceleration. It remains to find companies ready to commit to AI to take the torch from Henry Ford and make it a “Game Changer”.

READ: The first part of our article series: Can artificial intelligence be used in business?

The first articles in this series discuss the need for businesses to accelerate the adoption of artificial intelligence. Patrick Darmon will then propose a new approach to the successful adoption of AI in business. But before that, he'll return to the current paradigm of AI “scaling” to understand why it likely won't be the main driver of AI strategy for companies in the future.

Leave a Reply

Your email address will not be published. Required fields are marked *