In February 2024, the Cigref (1) published a note (2) aimed at sharing experiences and best practices in generative artificial intelligence (GenAI) in companies.
This initiative, spearheaded by the Cigref working group «AI & Generative AI», follows the rapid evolution of artificial intelligence technologies and their increasing adoption in the professional world, highlighting the importance of understanding and exploiting AI tools while managing their risks.
Here is our summary of Report by Cigref and the experience of returning Cyrille Chaudoit, Co-founder of talenco and expert in digital transformations, who supports many companies in their projects to adapt to generative AI.
Accompanying the growth and appropriation of generative artificial intelligence
For the Cigref, the rise in the power of generative artificial intelligence « requires one Acculturation of citizens and companies with data and AI tools. Training and awareness are essential for Se saisir des opportunités offered by AI, while recognizing its limitations and avoiding attributing to it human qualities« .
The report emphasizes the importance of maintain control over the quality of the data, the nature of the requests, and the results produced by AI tools.
Cyrille Chaudoit specifies: « In the current state of deployment of generative AI projects, we are witnessing a paradox: employees are embracing the integration of GenAI but fear for their jobs. This is due, in part, to our tendency to anthropomorphize all our technological tools. The associated terminology (intelligence, hallucinations, assistant, etc.) is no stranger to this. The Cigref is right to insist on this point. In this case, it is the human skills that need to be developed that must receive all our attention.«
« Employees overwhelmingly favor the integration of GenAI but fear for their jobs.«
Best practices for integrating artificial intelligence into the company
The Cigref working group discusses the importance of adopting a structured approach to identify the benefits and risks associated with generative AI. Managing and operating AI tools requires expertise. coherent approach integrating AI solutions within the company’s overall strategies.
The organizations participating in the working group shared 3 essential best practices, applicable to all types of businesses:
- Start with small-scale experiments and practice the « test and learn »
- Working with reliable partners
- Choosing the AI tools most suited to the company’s specific needs
Cyrille Chaudoit confirms the first point: « Most of the companies we support started by testing the potential of GenAI tools through the implementation of playgrounds secured environments allowing the various professions to experiment with use cases through POCs (Proof of Concept). In the field of AI, the test and learn It’s going very fast. The real challenge now is scaling up!« .
We mention this point in particular in our Webinar Generative artificial intelligence: how to support your teams? with François Debois, from L’Oréal.
Accompanying generative AI
Discover TALENCO our AI offer
Use cases for generative artificial intelligence tools
According to Cigref, companies must establish specific use cases and evaluate the available AI solutions and their roadmaps. Applications of generative AI are divided into three main categories according to the report:
- the improvement of internal productivity,
- the automation of operational processes,
- the creation of new services or products offering a competitive advantage.
Cyrille Chaudoit explains: « The use cases for generative AI are numerous. They obviously vary depending on sectors and professions. The first driver that usually comes to mind is the gain in productivity. But it goes without saying that before considering deploying an AI solution, it is necessary to examine its true usefulness and its long-term viability, as well as the reallocation of these productivity gains vs. the employees concerned. The HR function and Strategic workforce planning (upskilling/reskilling) must be a cornerstone of these reflections before moving on to the next level.« .
« The HR function and Strategic workforce planning They should be a cornerstone of AI-related reflections before moving on to a larger scale.«
What uses are there for generative artificial intelligence tools?
- Automate AI automates execution when the system has all the data necessary for decision-making.
- Make sound decisions AI helps with decision-making, but execution is ensured by the right skills within the organization.
- Recommend When AI does not have enough context to make a decision, it can issue recommendations to help with the decision-making process.
- Inspire AI offers insights and reflections based on the information available; HRs use it to produce, anticipate, and innovate.
- Evaluate The AI analyzes scenarios proposed by the teams.
- Generate AI generates different kinds of content.
- Teaching and advancing AI allows for recording entries and identifying how to progress.
Beyond the uses, several criteria such as data privacy, reliability of results and user experience They must also be considered when choosing a generative AI tool.
Generative AI technologies offer significant opportunities for innovation and improved efficiency. However, they must be adopted in an informed and responsible manner to maximize their benefits while minimizing potential risks.
Success in integrating artificial intelligence into the enterprise requires a rigorous approach, good data governance and a constant assessment of risks and benefits. It is also essential to Raising awareness and training the teams aux AI skills to ensure the proper appropriation and use of generative artificial intelligence.
Need to support management and teams on the impacts of artificial intelligence?
(1) Cigref is a network of large French companies and public administrations whose mission is to develop the capacity of its members to integrate and master the digital world. Through the quality of its thinking and the representativeness of its members, it is a unifying actor in the digital society. An association established under the 1901 law, Cigref does not carry out any lucrative activity.
(2) Note drawn from the reflections of the Cigref Working Group on «AI & Generative AI», led by Baladji SOUSSILANE, Vice President Digital & IT of the Air Liquide Group, a member of the Cigref, and moderated by Marine de Sury, Director of Mission at the Cigref, and bringing together more than fifty of its members to exchange on their practices and share their experiences.
