The digital sector in France is going through a period where European regulatory frameworks are catching up with the pace of innovation from tech players. Between the gradual rollout of the AI Act, the tightening of cybersecurity standards, and the changes in Google search rankings, the landscape has been reshaping at a rapid pace since the beginning of 2026. This article outlines the key facts and the questions they raise.
Cyber Resilience Act and NIS 2: the dual regulatory challenge of 2026
Tech media extensively cover artificial intelligence, but another regulatory front is mobilizing technical teams: the convergence between the Cyber Resilience Act (CRA) and the NIS 2 directive. These two European texts impose security obligations in distinct areas, which ultimately overlap.
The CRA targets manufacturers and publishers of digital products (software, connected devices, firmware). It requires documented management of vulnerabilities throughout the product lifecycle. NIS 2, on the other hand, addresses service operators and imposes incident notification and cybersecurity governance requirements.
For a company that publishes SaaS software used by a service operator, both texts apply simultaneously, with different reporting obligations. This overlap creates a concrete compliance issue that legal and technical departments must navigate. Field feedback varies on this point: some companies pool their compliance processes, while others treat each text in isolation due to the lack of harmonized guidelines.
To keep track of these developments over the weeks, news on opensyd.fr regularly covers the practical implications of these regulations for digital professionals.
AI Act: a staggered obligation timeline until 2028
The European regulation on artificial intelligence is not just a principle text. Its implementation follows a precise timeline that alters the roadmap for European tech companies based on the category of their systems.
The first explicit bans related to generative AI will come into effect as early as December 2026. These bans specifically concern the generation of non-consensual intimate images and the production of child sexual abuse material by AI. These prohibitions directly affect platforms, creative tool publishers, and content hosts.
For high-risk AI systems, deadlines have been rearranged. Heavy obligations (risk management, traceability of training data, technical documentation, post-market monitoring) will apply in several waves:
- December 2027 for high-risk AI systems operating autonomously
- August 2028 for AI systems integrated into existing physical or software products
- Enhanced transparency obligations for all generative AI models, without waiting for these final deadlines
This partial delay provides companies with time, but it also creates an area of uncertainty. The available data does not yet allow for measuring the actual compliance costs for SMEs and mid-sized digital companies. Large groups have already formed dedicated teams, while smaller organizations often discover these obligations gradually.
SEO trends and Google search in 2026: what is changing concretely
Natural referencing is undergoing a structural transformation with the rollout of Google’s AI Overviews. These AI-generated summaries appear at the top of search results and alter the visibility of traditional websites.
The phenomenon of “zero-click search” is increasing. Google provides answers directly in its results pages, reducing traffic to source sites. For content publishers, the question is no longer just about ranking on the first page, but about being cited as a source by Google’s AI.
Several SEO adjustments are emerging this year:
- Optimization for AI responses involves structured content, with direct answers in the first paragraphs that algorithms can easily extract
- User experience (Core Web Vitals) remains a ranking factor, but Google now incorporates signals related to the contextual relevance of content
- Long and generic content is losing ground to targeted articles that address specific search intents
- The Search Generative Experience pushes brands to rethink their digital marketing strategy, prioritizing brand awareness and thematic authority
However, the technical fundamentals of SEO (loading speed, semantic markup, internal linking) remain prerequisites. SEO in 2026 combines traditional optimization with adaptation to new AI search formats.
AI adoption in business: the problem is not the technology
Generative AI tools are available, accessible, and increasingly effective. The main barrier to their adoption in businesses lies elsewhere: in change management.
Field feedback identifies three recurring profiles in the deployment of AI within organizations: enthusiastic users who adopt tools without waiting for directives, resistors who openly express their reluctance, and a silent majority who do not use the tools but do not oppose them either. This silent majority represents the real challenge of transformation programs, as they escape the radar of project teams.
Traditional training programs (webinars, tutorials, internal champions) mainly reach the first two groups. The third remains largely outside the scope, limiting the return on investment of AI tool licenses deployed across an organization.
The organizational question is compounded by a regulatory challenge. With the AI Act, companies must document the use of their AI systems, train their teams, and establish governance. These requirements transform AI adoption from a purely technical subject into a cross-cutting project that involves legal, HR, and business departments.
The coming months will determine whether French companies can articulate regulatory compliance and operational transformation. The timeline of the AI Act imposes concrete deadlines, cybersecurity standards are tightening, and the rules of the game for SEO continue to evolve. Each of these topics deserves regular monitoring, as the details of application matter as much as the overarching principles.