The year 2024 has accelerated several technological transformations initiated in previous years, but it has also introduced a new element: a binding regulatory framework for artificial intelligence in Europe. Between the adoption of the AI Act, the rise of generative AI in businesses, and ongoing questions surrounding cybersecurity and sovereign cloud, the digital landscape of 2024 is not just a list of gadgets showcased at CES in Las Vegas.
This article revisits the driving forces that truly shape the market, beyond mere announcements.
AI Act: the European regulation reshaping tech roadmaps
The significant event of 2024 for the digital sector in Europe is not a product, but a piece of legislation. The Regulation (EU) 2024/1689, known as the AI Act, was published in the Official Journal on July 12, 2024, and came into effect on August 1, 2024. It is the first horizontal and comprehensive legislation governing artificial intelligence at the continental level.
The text classifies AI systems into four risk levels. Certain practices are outright banned: behavioral manipulation, exploitation of vulnerabilities, and some real-time biometric uses. Systems deemed “high risk” (automated recruitment, social scoring, assisted medical diagnosis) must comply with documentation, transparency, and human oversight obligations.
Analyses published on the tech section of Qui-Peut.Info show that this framework is concretely changing the priorities of technical management in France and Europe. Compliance with the AI Act is becoming a design constraint, not just a simple post-audit.
Since February 2025, bans and obligations for AI literacy are already applicable. Rules targeting large language models (LLMs) began to apply in August 2025. For companies developing or deploying AI solutions, the timeline is tight, and feedback from the field varies regarding the actual maturity of organizations in meeting these requirements.
Generative AI in business: between rapid adoption and poorly managed data
Generative AI dominated technological announcements in 2024. Major cloud solution providers (Microsoft, Google, Amazon) have integrated AI assistants into most of their professional suites. In France, players like Mistral AI have solidified their positioning in the language model market.
Adoption by businesses has been rapid, but feedback from the field varies on this point. Several questions remain open:
- The quality of training data used by internal models remains difficult to audit, complicating compliance with the AI Act for high-risk systems.
- The energy cost of training and inferring large language models poses a consistency problem with the sustainability commitments made by the same companies.
- The issue of intellectual property for generated content has not found a stabilized legal answer in 2024, neither in France nor at the European level.
The market enthusiasm for generative AI is real. The available data do not yet allow for conclusions on the actual return on investment for the majority of companies that have deployed it.
Cybersecurity and sovereign cloud: the blind spots of digital transformation
The proliferation of cloud solutions and AI tools has mechanically widened the attack surface of organizations. In 2024, cybersecurity remained a growing expense, driven by the increase in ransomware attacks targeting local authorities, healthcare facilities, and SMEs in France.
The sovereign cloud has become as much a political issue as a technical one. The SecNumCloud certified offerings, backed by ANSSI, aim to ensure that sensitive data remains hosted and processed on European territory. However, the number of actually certified offerings remains limited, and the cost compared to American hyperscalers hinders adoption by mid-sized organizations.
The specific case of healthcare
The healthcare sector illustrates this tension well. Digital innovations (telemedicine, digital twins, diagnostic AI) face particularly strict data protection requirements. Hosting health data in France remains subject to HDS certification, and the Health Data Hub projects have reignited the debate over the use of non-European cloud infrastructures.
Displays, connected objects, and mixed reality: what CES 2024 really changed
CES 2024 showcased several categories of products: transparent displays, AI-driven autonomous vehicles, and sensor-laden domestic robots. These spectacular demonstrations deserve to be placed in their market context.
Transparent displays remain confined to professional and event uses. Their production cost and limited brightness in real conditions hinder any short-term mass adoption. In contrast, advancements in mixed reality (headsets combining virtual and augmented reality) have gained credibility with the arrival of new products positioned in the professional market.
On the connected objects side, the 2024 trend is less about the proliferation of devices and more about their interoperability. The Matter protocol, supported by Apple, Google, and Amazon, is slowly progressing but could simplify the connected home ecosystem in the coming years. The available data do not allow for conclusions on the actual adoption rate by third-party manufacturers.
Sustainable digital: a discourse struggling to translate into indicators
The notion of “green IT” has been omnipresent in the communications of tech companies in 2024. Rethinking the entire lifecycle of digital equipment (manufacturing, usage, end of life) is presented as a priority by most major groups.
Public data on the actual carbon footprint of the digital sector remains fragmented. In France, ADEME and Arcep have published measurement methodologies, but their adoption by companies is still partial. The gap between stated commitments and measurable practices constitutes one of the blind spots in the discourse on digital innovations in 2024.
The topic of refurbishment is progressing, driven by specialized players and the anti-waste law (AGEC law). Refurbished volumes are increasing in the smartphone market, but the market share remains modest compared to new devices.
Thus, the year 2024 will have been one where technology faced regulatory, energy, and security constraints head-on. The most visible innovations are not necessarily those that restructure the market: the AI Act, cybersecurity, and data issues weigh more heavily on companies’ decisions than the latest transparent display presented in Las Vegas.