After more than six months of work, on March 28, 2018, mathematician Cédric Villani made public his report on artificial intelligence, the culmination of the parliamentary mission entrusted to him by the Prime Minister in September 2017.
The starting point of this report is simple: without the Big Data phenomenon, artificial intelligence would not be able to achieve the economic and societal success that has characterized it in recent years.
Historically, it was the scientific and military communities that first conducted research on artificial intelligence in the 1950s. Some computer manufacturers have contributed to the media coverage of the cognitive capacities of their artificial intelligence systems, for example in the context of marketing operations, such as IBM with the six chess games played in the 1990s between the super-computer Deep Blue and Garry Kasparov, the world champion at the time.
Datafication was made possible by the creation of the Internet, by the multiplication of computing capacities of consumer computers and by GAFA*, the first groups to have based their business models on data that was initially raw, then restructured, and finally made meaningful, and therefore able to be economically valued.
It is this new black gold that many start-ups have been refining for several years to extract added value, with varying degrees of success. However, any company, recent or not, has an economic and notoriety interest to engage a reflection and strategic decisions around an innovation based on artificial intelligence.
Without Big Data, there is no AI
The Villani report first points out the importance of the size of the corpus of available data in order to use artificial intelligence. Raw, unanalyzed data has very little value in itself. It is only when it is connected to other data in data lakes that it can be made to speak, to produce knowledge, and to draw optimized and therefore marketable information.
Companies like Ellisphere, which have been handling large databases of structured data for many years, have developed unique expertise and depth of analysis. To participate in the artificial intelligence movement, this experience must be complemented by an essential part of it: deep learning, or machine learning. What a human cannot do because of the repetitive nature of certain tasks or because the volume of data is too large, an algorithm can do by correlating a very large amount of data.
Artificial intelligence for profitability
The partial reproduction of the cognitive capacities of the human being allows the optimization of the numerical data present in the large databases of certain companies. Thus, an algorithm trained for this purpose can learn from the data it manipulates, and draw new information from it.
This artificial intelligence, which could also be described as superficial (considered by specialists as still being of low level), nevertheless has the advantage of allowing a system to create new information, and therefore added value, in an automated and continuous manner.
The combination of relevant algorithmic models and structured data can thus be a major growth driver, regardless of the size of the companies involved in an innovation movement linked to artificial intelligence. This is the profitability aspect of artificial intelligence highlighted by the Villani report.
An enhanced reputation
The Villani report also emphasizes the value of the company's communication policy around artificial intelligence. The gain in reputation linked to the development of operational solutions based on artificial intelligence is an asset.
The Villani report recommends two approaches: the creation of labels and the establishment of prizes for innovation in the field.
As for labels, the idea is to build on the French Tech label, whose success in just a few years has been to federate, network and make the French digital ecosystem visible. Thus, a new AI label is envisaged to promote companies that optimize their data thanks to artificial intelligence. Informal meetings (meetups), conferences, and company communications could be organized within this new ecosystem, as part of this label.
The field innovation awards could highlight companies, start-ups or not, that develop innovative operational solutions that have given satisfaction to their customers.
This part of the Villani report states the potential benefits of AI in terms of notoriety.
Cooperative innovation encouraged
Finally, any creative movement is not sustainable if it remains static and limited. The Villani report talks about the need for "agile and diffusive" research. It therefore proposes the creation of Interdisciplinary Institutes of Artificial Intelligence (3IA), which would bring together researchers, students and companies.
Companies could thus interact regularly with researchers on their business problems, and thus develop their solutions according to the advice they receive. These exchanges could, for example, concern the optimization of their algorithms or audits of their AI systems.
Private entities could also be associated with the files presented by the institutes, as affiliated members (for example, through an annual fee adapted to the company's financial resources).
This interaction between companies and research would boost theinnovation movement and the emulation linked to artificial intelligence.
Recommendations already planned for implementation
These recommendations of the Villani report, stated under the conditional tense, were for some of them clarified and sometimes rephrased in the future tense by the President of the Republic, Emanuel Macron, on March 29, 2018, during his speech at the Collège de France dedicated to the presentation of France's plan for the development of artificial intelligence.
It was also announced on this occasion that the "porosity between the public and private sectors" should be encouraged , with, for example, researchers who should be "authorized to spend half their time in a private entity, whereas today it is 20%".
The President also stated his desire to set up a network of four or five interdisciplinary artificial intelligence institutes whose vocation will be to accept private funding.
It is certain that many of the recommendations made in the Villani report will be the subject of political decisions in the coming months.
Innovation, profitability and notoriety can give full meaning to the potentially considerable development of artificial intelligence in our society, and more particularly within companies that, like Ellisphere, have chosen innovation to optimize and enhance their operational solutions.
*GAFA: Google, Apple, Facebook, Amazon
