I am a serial entrepreneur, and generative AI reaching maturity has generated a lot of enthusiasm, as the applications of this technological revolution are truly significant and far-reaching.
The launch of ChatGPT coincided with a break in my career, after the sale of my previous company. I explored this innovation without immediately bringing it into a professional setting.
After my break, I assumed that businesses would have taken the subject up. Sounding out the people around me, I found that although some share my enthusiasm and have become converts, most use it very little, or not at all. Here are the main reasons.
Resistance to change
Like any transformation, generative AI meets with a certain resistance. This innovation particularly affects intellectual professions, which previous technological revolutions often spared.
The resistance takes different forms: some point to errors (hallucinations) to conclude that the technology is not reliable, others raise philosophical questions, and some refuse to accept that a machine can carry out intellectual tasks they used to do themselves.
Security
The new technologies do not yet offer sufficient guarantees on data handling. Many questions arise about how data is processed, whether it is used to train models, and the risk of leaks. These concerns are legitimate for chief information officers. In addition, many companies are looking for sovereign solutions for reasons of compliance, control and geopolitics.
Accessibility
These new technologies are far more accessible than earlier ones. All it takes is a simple request to get an answer. It is not as simple as it looks, however, as the trend around prompt engineering shows. How the instruction is worded is crucial.
There are similarities with managing a junior assistant: to get convincing results, you have to be as clear as possible and check the work. Although the chat model has become the norm, new user interfaces (UX) are very likely to appear to support these interactions and make them easier to use in certain projects. We firmly believe in this evolution and are actively working on it.
A dizzying pace of change
The technological advances in generative AI are impressive and fast. This dizzying pace poses challenges of adaptation and implementation, but it also offers unprecedented opportunities to transform many sectors.
In just one year, generative AI has developed quite spectacularly. After OpenAI’s phenomenal launch of ChatGPT, an avalanche of announcements followed, from the big tech companies and from new start-ups, including French ones such as Mistral AI. These technologies do not just generate high-quality text: they also produce images and videos, and can even imitate voices.
This flurry of innovation can unsettle IT leaders and decision-makers, who tell us how hard it is to choose which technology to invest in. Many prefer to wait for the market to settle before making a decision.
The uControlAi project
To shape my latest entrepreneurial project, I focused on four main issues. The aim is to develop a solution that meets the expectations of IT leaders by securing and framing usage, while offering access to private AI to preserve data sovereignty. Within a few months, I also realised that it was possible to offer new interfaces for an even more inclusive experience, making collaboration between employees easier.
However, refusing to master these tools means giving up technologies that can revolutionise the way we work. This is a major issue for each individual, for companies and for the economy. For an individual, an employee who does not take on these advances risks becoming less employable. For a company, it would mean a loss of competitiveness with certain economic consequences.