Saving one day a week thanks to AI: myth or reality?

Can you tell us about your career path and
Explain how you identified this need.

I have a background in business and began my career in the United States in export sales positions before moving into e-commerce, first at Decathlon and then in Switzerland at Visilab (now Essilor Luxottica), where I was in charge of omni-channel strategic marketing after leading various digital projects across the group and supervising customer service.

At the same time, I have been training professionals in digital marketing for about ten years.
 

At the end of 2022, the arrival of Chat GPT awakened my geeky side, and I quickly got my hands on generative AI to form my own opinion.

Very quickly, when I tried out a feature, I found myself thinking, “My goodness, if only I'd had this a few years ago when I was doing that task, I could have sorted it out in a matter of hours instead of weeks!”.

This led to the project of training professionals in artificial intelligence, using my own experience to distinguish between AI gadgets and applications with high added value for professionals.

What technical and interpersonal skills are most sought after in the AI sector today, and what advice would you give to our learners who wish to pursue a career in this field?

Professional expertise: For me, what makes a professional capable of
Generating value with AI is above all its business expertise.

Indeed, it is thanks to this that a professional will be able to identify the most interesting use cases to implement in their field.

This is generally something that will be very difficult for someone who is not proficient in the job.

Let's imagine that you want to create a neurosurgeon robot. Let's assume that you are capable of easily creating a robot that can do anything, provided that you explain it clearly.

If you are not a surgeon yourself, it will be very difficult to provide these
explanations. This is what is happening with AI today.

Critical thinking: AI regularly makes mistakes, whether it is a hallucination
specific to the tool or a correct answer to a question that has been poorly phrased by the user.

Today, this makes control a very important step in the use of AI. This control is made possible not only by professional expertise, as I would like to point out, but also by a critical mind that the pre-ChatGPT generations were fortunate enough to be able to develop.

This critical spirit is as essential as it is fragile.
Therefore, it is important to continue developing and using it.

Mastery of tools: Strangely enough, this is probably the easiest part. These tools are very intuitive thanks to natural language, even though new ones are constantly appearing and surpassing the others. Today, there is less and less demand for advanced technical skills in a specific tool, which may quickly become obsolete. 

On the other hand, knowing how to work with generative AI tools in general is becoming a skill in the same way as knowing how to use a computer. 

There is also a reason why a foundation of training in the use of AI in the workplace is essential today.

In fact, in the early hours of the training courses I give, I observe learners who are highly skilled in their profession but completely inhibited when they have to perform tasks on a tool or in an environment they are not familiar with.

It is only when people feel sufficiently comfortable with this new technology that they begin to fully utilise the two skills mentioned above.

Could you describe a specific project in which you implemented an AI solution, detailing the stages of the project, the challenges encountered and the results achieved?

For reasons of confidentiality, I cannot necessarily describe the most
important, but they all follow a common pattern (described below). 

Paradoxically, the most successful projects are the simplest team-building exercises.

Indeed, among the quick wins that companies can exploit, “team empowerment” is, in my opinion, the most profitable at present.

The vast majority of a company's employees can very quickly be trained in general AI tools (such as Chat GPT Claude Mistral or others) and become aware of the enormous potential they can easily achieve with them. 

The time savings and efficiency gains are therefore enormous, although difficult to measure as they vary greatly.

But the idea of saving one day a week is far from crazy.

Everyone can then easily calculate the savings this represents!

How can non-tech companies approach
the integration of AI into their processes? What are the common mistakes to avoid and best practices to adopt?

The first step involves training the decision-maker(s). Equipped with a visual
The clearer they are, the better they are able to formalise precise briefs for their teams, while being aware of the prerequisites necessary for the successful implementation of their project.

The next step is to form teams according to the programme defined with the decision-makers.

Once the level of knowledge and understanding has been standardised, the next step is to choose the tools and make the company data available to feed into the chosen AI solution.

The implementation, whether internalised or outsourced, depends on
the organisation.

Then finally comes the testing phase on low-risk cases for training and
continuous improvement of the solution.

For the most complex organisations, an internal launch phase will be
also necessary in good change management practice.

Among the challenges encountered are often obstacles that can arise when leading a project within an organisation, namely: misaligned objectives, poor internal communication, and a mismatch between ambitions and available resources.

All these points lead me to believe that the key to the success of an AI integration project lies in effective human resource management.

What is your vision for the evolution of AI over the next five years, and what new professional opportunities could emerge for our learners?

Ah! The acceleration curve for these technologies is truly exponential! In my training courses, I incorporate tools that are sometimes less than two weeks old, so it's very difficult to predict what will happen in five years' time... 

But with the saturation of digital tools, the optimist in me sees a revival on the horizon.
the importance of everything that makes us human, namely: face-to-face contact, the importance of local networks, the search for authenticity and opportunities to disconnect. 

Similarly, among the medium-term challenges I see for the
Companies will face the challenge of human resource management with the risk of an explosion of
burnouts.

Not because of an increase in workload, but rather because of an acceleration in pace that our bodies are finding increasingly difficult to cope with. 

To conclude, I would say that one of the key areas of focus for tomorrow's managers must be to properly address the following question: If my teams are able to perform their tasks 50% faster, what should be done with the remaining 50%? 

To help us with this, we can always ask AI for a (little) help!