A chatbot ranked at the same level as a major social network by the European Commission, transparency obligations that change the way we consume content online, a French regulation that now treats AI services as media: the current trends are not just headlines. They are reshaping the rules of the digital game, and it is beneficial to understand what is happening behind the scenes.
ChatGPT, Reddit, and Roblox under the enhanced regime of the European DSA
At the end of August 2026, the European Union placed ChatGPT, Reddit, and Roblox under the strictest obligations of the Digital Services Act (DSA), after these platforms crossed the threshold for classification as very large platforms. In practice, this means mandatory risk assessments, increased transparency on recommendation algorithms, and strengthened measures for the protection of minors.
What stands out is the case of ChatGPT. It is the first artificial intelligence chatbot explicitly classified at this level in the EU. Until now, the DSA primarily targeted social networks and marketplaces. By integrating a conversational service, the Commission acknowledges that interactions with generative AI pose risks comparable to those of a social platform with tens of millions of users.
To keep up with these regulatory developments and other topics shaping digital news, you can access news on Au Top and cross-reference sources before forming an opinion.
Transparency obligations of the AI Act: what changes concretely online
Since August 2, 2026, the transparency obligations set out by the AI Act have come into effect in the European Union. The principle is straightforward: every conversational system must inform the user that they are interacting with artificial intelligence. Content generated or modified by AI must be labeled as such.
On the ground, this affects customer service chatbots, automatic writing tools, and image generators. AI-generated content must now carry an explicit label. This is no longer a recommendation: it is a legal obligation, with possible penalties.
The impact for content creators and online media is tangible. An article, image, or video generated by AI circulating on a European social network without labeling violates the regulation. Platforms hosting this content become responsible for implementing the reporting.
Why this rule changes the game for media and social networks
Until now, distinguishing between human-written text and AI-generated text relied on personal judgment. With this obligation, we move to a normative framework. Platforms must adapt their interfaces, and users have a reliable reference point. Feedback varies on how each service implements this obligation, but the legal framework is established.
French AI regulation: an unprecedented audience barometer
The French regulatory authority for communications and audiovisual has launched in the first half of 2026 a barometer of AI service audience. This is a strong signal: France is beginning to treat artificial intelligence services as full-fledged media actors, worthy of being measured and regulated just like a television channel or a radio station.
This barometer allows for quantifying the actual use of AI tools by the French public. We move from rough estimates to a structured measurement logic, comparable to what exists for the audience of websites or audiovisual programs.
What this barometer implies for digital players in France
The consequences are multiple:
- AI services become comparable to each other based on public audience criteria, altering the competitive dynamics
- Advertisers have reliable data to guide their advertising investments towards the most used AI tools
- Regulators can adjust their requirements based on actual audience, not just the number of registered users
This approach positions France among the first countries to structure the measurement of AI audience as a regulatory tool, not just as a statistical curiosity.
Online news: how to filter out the noise and spot underlying trends
In the face of the volume of information published every day between continuous news feeds, push alerts, and algorithmic recommendations, we often confuse urgency with importance. Underlying trends are rarely spotted in notifications.
Here are some concrete criteria to help sort through:
- Check if information has measurable regulatory or economic consequences, not just viral potential
- Cross-reference at least two independent sources before considering a fact as established
- Distinguish institutional announcements (laws, regulatory decisions) from the surrounding opinion comments
- Monitor European decisions on digital matters, which directly impact the services used daily in France
Algorithmic rankings on Google News or social networks prioritize freshness and engagement. This is not a flaw, but it creates a bias towards event-driven content at the expense of structural developments. Regulatory changes like the DSA or the AI Act often go unnoticed while they permanently alter the digital environment.
Online monitoring: choose your channels rather than follow everything
Multiplying sources does not improve understanding if filtering is not applied. A well-configured aggregator on three or four specific themes (digital regulation, European economy, technological innovation) provides more useful monitoring than a general news feed consulted continuously. The quality of monitoring depends on sorting, not volume.
The current trends in the digital space, whether it is the ranking of ChatGPT under the DSA, the mandatory labeling of AI content, or the audience measurement of artificial intelligence services in France, share a common point: they make the regulatory framework more readable for the end user. We are moving from a web where rules were vague to an environment where obligations are explicit and penalties possible. Staying informed on these topics means keeping control over how we consume and produce content online.
