ECONOMY – State sovereignty and regulation of the digital economy : The case of digital regulation in Germany and France

By François Souty

Executive Summary

The digital economy initially seemed to enshrine the erasure of national sovereignty in the face of the power of global platforms capable of deploying their economic models on a global scale. The European movement that has been underway since the early 2000s has gradually reversed this perspective: from traditional competition law to the Digital Markets Act (DMA), then to the Artificial Intelligence Act (AI Act), the European Union has undertaken to regain a capacity to regulate the digital economy ex ante. However, this Europeanisation does not eliminate national differences. The comparison between Germany and France shows that the same technologies can be used in very different economic and administrative environments. The case of Uber is particularly revealing: Germany tends to gradually integrate the platform into a traditionally regulated sector, to the point of allowing taxis themselves to benefit from the digital tool; France has built a set of successive professional, social and administrative rules around ride-hailing services and platforms. This difference reflects two conceptions of the regulatory state. German regulation appears to be more subtle, evolving and attentive to the adaptation of SMEs and the Mittelstand to economic transformation; French regulation, without being devoid of justification, appears more abundant and prescriptive, with the risk of a disproportionate compliance burden for small companies. A large global platform can absorb this complexity; an innovative SME much less easily. The challenge of the AI Act will therefore be to achieve European harmonisation that protects citizens and markets without turning regulatory compliance into a barrier to innovation. Digital sovereignty is therefore not disappearing with Europe: it is moving towards the ability to define, apply and adapt the rules that concretely organise the digital economy.

Introduction

A little over a quarter of a century ago, the digital economy seemed to herald a new geography of power. National borders seemed to be losing some of their reach in the face of companies capable of instantly offering the same services to hundreds of millions of users, moving their capital and infrastructure, collecting data on a global scale and organizing markets whose center of gravity was largely beyond the control of national administrations. Google, Amazon, Apple, Microsoft, then mobility platforms, social networks and new players in artificial intelligence thus gave shape to an economy whose economic power seemed inversely proportional to its territorial position.

Recent history has shown, however, that this apparent deterritorialization does not mean the disappearance of the state. Rather, it has led to a transformation of its instruments. After having long dealt with digital companies by means of common competition law, the European Union has gradually developed a real digital regulation policy. The DMA, data regulation, cybersecurity measures and, now, the AI Act bear witness to the same evolution: the state and the Union are no longer content to sanction after the fact behaviors that they deem anticompetitive or dangerous; they seek to define in advance the conditions under which certain markets and technologies can operate.¹

This transformation is an extension of an earlier evolution of the economic state. The liberal state of the nineteenth century was mainly intended to guarantee the general rules of the market. The industrial state of the twentieth century has become a producer, investor, planner or protector. The regulatory state that has been developing since the last decades of the twentieth century acts differently: it does not necessarily own companies, but determines the conditions under which they can carry out their activity. The digital economy pushes this logic to its conclusion, since the rule can now relate not only to the company or the market, but to the algorithm that organizes the meeting between supply and demand.

Our previous analyses on the DMA, artificial intelligence, cybercrime and the European Commission’s policy towards Google have made it possible to observe the different dimensions of this change.² They have also shown that European digital sovereignty cannot be understood as a simple transfer of competences from States to Brussels. The European standard can harmonise the principles without fully standardising the economic conditions for their application.

This is precisely what the comparison between Germany and France reveals. Two Member States subject to the same major European regulations, belonging to the same internal market and confronted with the same platforms can nevertheless organise their digital economy differently. Legal traditions, industrial structures, administrative organisation, industrial relations and economic policy choices continue to leave their mark on the concrete functioning of markets. And the overall economic performance of each of the nations ultimately appears to be highly differentiated, as can be seen strikingly in 2026.

The case of Uber offers a privileged observatory in this regard. The platform combines almost all the dimensions of the digital revolution: geolocation, data, networking algorithms, dynamic pricing, platform economy and transformation of work. But at the same time, it encounters an old sector, that of individual passenger transport, which is heavily regulated by the state. The confrontation between Uber and taxis is therefore not just a quarrel between a technology company and a traditional profession. It is a life-size experiment in how a state can choose to welcome, channel or contain digital innovation.

Through Uber, two conceptions of regulation are thus emerging. Germany seems to be more interested in gradually adapting existing structures to new technology, while France tends more willingly to superimpose new prescriptions on existing ones. The difference is particularly important for SMEs: complex regulations can be absorbed by a large platform with powerful legal and technical services, but become a real fixed cost of entry and expansion for a small company.

The question is therefore less to know which country regulates the most than to determine which regulation best allows a State to simultaneously preserve competition, innovation, consumer protection and the capacity of its companies to grow. It is by this yardstick that digital sovereignty must now be assessed.

I. A digital economy subject to different regulatory sovereignty

The rise of European digital regulation is inseparable from a realization: traditional competition law was not designed for markets in which network effects, data, economies of scale and marginal costs close to zero allow a player to consolidate a dominant position very quickly. The problem that emerged with the large platforms was no longer just that of the abusive behavior of an already dominant company; it was that of the very possibility of preserving the contestability of the market before this dominance became practically irreversible.

The DMA is the most ambitious response to this difficulty to date. It marks the transition from an essentially ex post logic to an ex ante logic. It is no longer a question of waiting for abuse to be demonstrated at the end of an often lengthy procedure, but of imposing obligations on certain systemic platforms designed to prevent certain practices in advance. This development is particularly significant in the case of Google, where the history of European procedures has illustrated the possible gap between the duration of competition law and the speed of transformation of digital markets.³

While the assessment of the effectiveness of the DMA is still largely premature because the text is too recent, the regulation of artificial intelligence is nevertheless going a step further. The problem is no longer just the economic power of a company, but the ability of an algorithmic device to produce decisions with economic or social consequences. The algorithm can determine the order in which information is displayed, propose a price, assign a risk probability, select a candidate, guide a consumer or offer a ride to a driver. A technology that seems abstract and mathematical thus becomes an invisible infrastructure of the market.

The history of economic regulation leads to an essential conclusion: when the state regulates the algorithm, it indirectly regulates the market. Digital sovereignty cannot therefore be reduced to the possession of an IT industry or the localization of data. It includes an institutional dimension: the ability to determine the rules according to which new technologies meet consumers, businesses and workers. Cybersecurity had already demonstrated this in the field of critical infrastructure; artificial intelligence now confirms it in the daily functioning of the economy.⁴

But this sovereignty is no longer exclusively national. The internal market and European regulations have transferred a substantial part of the normative power to the Union. The originality of the contemporary situation lies precisely in this superposition: increasingly dense European regulation meets States that retain their administrative traditions, their sectoral regulations and part of their economic policies.

This results in a paradoxical situation. The European Union can impose the same obligation on a platform operating in Germany and France, while the economic conditions under which this platform operates may remain different. European law tends to standardize the standard; the State continues to shape the market and its structures.

This issue is particularly important for SMEs. The burden of a regulation is never measured only by its legal content. It depends on the company’s ability to assimilate it. A large international platform may employ lawyers specialising in competition law, data protection specialists, algorithmic compliance engineers and economists who can measure the consequences. An innovative SME very rarely has such means. An identical obligation can therefore have radically different competitive effects depending on the size of the company.

Regulation can thus produce an unexpected paradox: by seeking to limit the advantage of large platforms, it can contribute to strengthening their relative position if the cost of compliance becomes a barrier to entry for smaller companies. The problem is therefore not the regulation itself, but the economic quality of the regulation, the way in which it has been developed.

Germany offers a particularly interesting field in this respect. The country is by no means an area of weak regulation. The administrative density and the weight of bureaucracy are regularly denounced by German companies themselves. But the tradition of the Mittelstand, decentralisation and the weight of the Länder have fostered a culture of experimentation, local adaptation and cooperation with companies. The policy of digitising SMEs is explicitly in line with this logic.⁵

France, for its part, has a different state tradition. The central state has historically played a much more direct role in the organization of economic sectors. This capacity is often a strength: it makes it possible to quickly impose a national strategy, to mobilize the administration and to create powerful mechanisms. But it also carries a symmetrical risk: that of responding to each economic or social difficulty with a new normative layer, most often to the detriment of SMEs, often with public indifference to the latter’s difficulties.

The European Commission still has a significant administrative and regulatory burden in France today, as well as a digitalisation of SMEs that remains uneven.⁶ The question is therefore not to reproach France for regulating, nor to present Germany as a model of deregulation. It is to compare two ways of using the regulatory power of the State.

The Uber case makes it possible to move from this general comparison to concrete observation.

II. Uber, a laboratory for differentiated digital regulation in Germany and France

Uber was born from the meeting of a relatively simple technology — a geolocation and networking application — and an old market whose organization was largely based on administrative rules: that of individual passenger transport. This encounter largely explains the conflicts that the platform has caused in Europe. Uber has not only introduced a new company; it has introduced a new way of organizing a market.

In France, the conflict quickly crystallized around the distinction between taxis and VTCs. Taxis benefit from a special regime, particularly for marauding, while VTCs can compete with taxis on the basis of prior reservation.⁷ Uber’s development has therefore created a new competitive market without removing the historical prerogatives of taxis, but also by radically distorting the system of access to markets and the financing of the sector’s pension system (taxis have seen the price of taxi plates collapse over years and ensuring a retirement capital at the end of the market 40 years later; in return, Uber does not assume the financing of the pension schemes of its self-employed drivers whose pensions will essentially be paid for… of the State budget or of third-party pension schemes that they will have financed little or nothing).

This model profoundly changed the economics of urban mobility. Consumers could now order a vehicle without knowing its location, compare waiting times, know in advance certain characteristics of the journey and, above all, be put in touch by an algorithm with an available driver. The progress was not only technological: it reduced the costs of research and coordination that had long structured the taxi market.

Germany has encountered the same innovation but in a different legal environment. In particular, the Personenbeförderungsgesetz maintains specific rules for chauffeur-driven rental cars, including the obligation to return to the headquarters when no new rides are received. This rule, which may seem anachronistic in the instant economy of the platform, nevertheless has a specific economic significance: it limits the possibilities of repositioning a vehicle and can increase the kilometers traveled without passengers.⁸

But the German development is interesting precisely because it has not remained fixed around this protective logic. The legislator has gradually introduced possibilities for adaptation to take into account the development of digital booking and local specificities. Municipalities thus have room for manoeuvre to organise the transport offer and regulate the conditions of competition.

This flexibility contrasts with the image often given of Germany as simply more restrictive than a supposedly more liberal France towards Uber. The reality is more subtle. Germany has sought more to bring the platform into the existing system, rather than to let the platform constitute a fully autonomous competing system.

This logic has become particularly visible with the opening of Uber to German taxis. Since 2024, the platform has been offering taxis throughout the country alongside the transport services offered by partner companies. Uber reported in 2025 that more than 10,000 German taxi drivers could be booked through it.⁹

The phenomenon is economically significant. The platform then ceases to be simply the instrument of new competition against taxis; it becomes a digital infrastructure from which taxis themselves can benefit. The conflict between innovation and traditional sector is turning into an integration process.

The pricing policy provides a second illustration. Several German cities have experimented with early and flexible pricing for taxis booked through apps. In particular, Uber claims to have seen a significant increase in the turnover of taxis using these devices.¹⁰ While this data should naturally be viewed with due caution when it comes from the company itself, it sheds light on a regulatory philosophy: the technology can be used to modernize the protected sector instead of just competing with it.

France has followed a different trajectory. The taxi retains its special regime while the VTC market has developed as a distinct market, largely organized around platforms. This duality has an obvious advantage: it has allowed the rapid emergence of a new offer and additional competition. But it has also created a complex regulatory system in which transport law, social law, competition law, professional rules and platform-specific regulations are superimposed.

Algorithmic pricing is a perfect illustration of this difference. The price of a ride-hailing trip can be adjusted according to the demand and availability of drivers. The algorithm thus becomes a market mechanism in its own right. It is no longer satisfied with organizing the meeting between two parties; it participates in determining the economic value of the transaction.

This innovation obviously raises legitimate problems: price transparency, consumer information, driver compensation, income predictability and possible algorithmic discrimination. But it also raises a more general economic policy question: to what extent should the state intervene in the operation of a digital price mechanism that precisely allows supply to be adjusted to demand?

France has gradually responded to these questions by densifying the social and professional framework of platforms. The social dialogue of platform workers, the rules relating to the conditions for exercising ride-hailing services, the fight against fraud and the new obligations imposed on players in the sector testify to a growing desire to regain control of a market whose initial functioning was largely based on the technology and self-regulation of the platform.¹¹

This intervention can of course be justified. But it poses a difficulty that the German comparison makes clear: at what point does the protection of the market and its players turn into a complexity likely to reduce its effectiveness?

The question is all the more important as the new obligations do not have the same effect on all operators. Uber has a global technological infrastructure. It can modify its application, develop new compliance tools, absorb legal costs and adapt its contracts. A French mobility SME does not have this capacity. The same regulatory obligation can therefore indirectly reinforce the advantage of a global player that it initially sought to regulate. Mutatis mutandis, the same disadvantage for SMEs is found in the field of distribution vis-à-vis Amazon platforms and other sectors where multinational digital systems operate.

The Autorité de la concurrence itself highlighted this risk when it examined, in 2025, an agreement relating to the freedom of choice of rides by VTC drivers. In particular, the Autorité noted the need to assess the economic and competitive effects of a system whose technological implementation was largely based on the capabilities of a large platform.¹²

The situation then becomes paradoxical: a public policy aimed at improving the situation of workers may require technological investments that only the largest platforms are able to make quickly. Social regulation can thus produce an indirect competitive effect.

German law seems, in several areas, to have sought more of a way of integration and simplification. The platform can become an instrument for modernising the taxi, while the rules can be adapted locally to market conditions. It is not a question of abolishing the regulations, but of making them flexible enough for companies to adapt to them. The contrast can be summarised as follows:

However, it would be wrong to infer from this that the German consumer always benefits from lower prices, that the German driver necessarily has a higher income or that Uber would be mechanically more profitable in Germany or France. Differences in taxation, wages, urban density, vehicle costs and demand structure make it impossible to make a gross comparison of the platform’s profits.

In this respect, the economic interest of the Uber case lies elsewhere. It allows us to observe how a rule modifies market incentives. The obligation to return to headquarters can increase empty kilometres; dynamic pricing can bring supply and demand closer together; advance pricing can improve predictability for the consumer; the integration of taxis into a platform can increase their usage rate; the accumulation of social rules can improve driver protection but increase the fixed cost of entry for new operators.

It is therefore the economic quality of regulation, and not its quantity, that must be measured. In this respect, the Franco-German difference becomes even more interesting. Germany seems to favour more readily an evolving, experimental and differentiated regulation, which seeks to gradually introduce innovation into existing economic structures. France seems to respond more to successive problems by adding new prescriptions. This trend is not unique to the VTC sector; it refers to a more general tradition of the French State, which has historically been inclined to produce the rule as a privileged instrument of economic and social transformation.

Of course, in the past, this tradition has produced great successes. It has enabled France to lead rapid economic transformations, to create large industrial sectors and to effectively mobilise the administration. But it now has a particular weakness: in the digital economy, the speed of innovation is such that there is a risk that regulations will constantly come after technology, and then pile on top of the previous rules.

Germany is not free of bureaucracy, far from it. But the Mittelstand‘s very pro-SME culture and the economic weight of SMEs have given particular importance to the issue of regulatory burden. The digital transformation policy of German SMEs has thus been accompanied by programmes aimed precisely at reducing the obstacles to the adoption of new technologies.¹³

The issue takes on strategic significance. A large global platform can live with regulatory complexity; an SME that seeks to become the next major European platform may be prevented from doing so. Overly complex regulation can therefore unintentionally protect acquired positions against new entrants. The real question for the strategic state is therefore not how to regulate more, but how to regulate better.

III. Towards European digital sovereignty? Gradual harmonisation through the AI Act

Uber’s experience reveals the limits of incomplete European harmonisation. As long as the Union does not fully regulate the conditions under which a sector operates, States continue to shape markets by their own choices. The transition from traditional competition to the DMA and then to the AI Act corresponds precisely to the European desire to reduce this fragmentation in areas where digital technologies have become structuring.

The AI Act is particularly revealing because it no longer targets only dominant companies. It seeks to directly organize the uses of artificial intelligence according to the risks they present. The regulation prohibits certain practices, establishes obligations for general-purpose AI models, and imposes specific requirements on high-risk systems.¹⁴

Its timing reflects a gradual ramp-up. The regulation entered into force on 1 August 2024 and started to take effect on 2 February 2025 with the prohibitions and obligations on AI culture. The provisions on general-purpose AI models and governance became applicable on 2 August 2025. The general framework and transparency obligations entered their main phase of application on 2 August 2026. The rules on certain high-risk systems are still to enter into force from 2027 and 2028.¹⁵

The historical significance of this timetable should not be underestimated. With the AI Act, Europe is trying to build a real common regulatory space for artificial intelligence before technologies are fully stabilised. This is a different approach from the one that characterised the first period of the digital economy, when platforms were able to achieve considerable economic power before public authorities had the necessary instruments to regulate them. The aim is therefore to make regulation a preventive instrument of European sovereignty.

But this sovereignty remains shared. The European regulation requires national authorities capable of applying it, receiving complaints, monitoring companies and supporting economic actors. Germany has thus organised an administrative architecture in which the Bundesnetzagentur has played a central role in the national application of the AI Act since 2026.¹⁶ France has adopted a different logic. It has not favoured a single national authority, but a complex decentralised organisation, multiplying the administrations and agencies responsible, calling on administrations and regulators already competent in their respective sectors. The DGCCRF coordinates them, in conjunction with the DGE, while the CNIL, Arcom, ACPR and sectoral authorities intervene according to the nature of the AI systems concerned; the ANSSI and the Digital Regulation Expertise Centre (PEReN) provide a shared base of technical expertise, i.e. a potential of a dozen administrations and agencies concerned resulting from the classic administrative layer.¹⁷

The difference is significant. In both countries, digital sovereignty is now exercised through the national application of a common European standard; but it is not organised according to the same administrative philosophy. Germany tends to concentrate and structure the regulatory function around a federal architecture adapted to its institutional organisation; France favours the coordinated mobilisation of its existing authorities, without merging, simplifying or reorganising. In either case, European harmonisation does not therefore eliminate national choices : it shifts sovereignty to the way in which each State organises control, expertise, support for companies and, ultimately, the effective implementation of the European rule. Here again, the European standard does not erase the State. It transforms its function.

The State is therefore no longer necessarily the main producer of the rule; it becomes its interpreter, its controller, its mediator and its accompanist. This evolution could even constitute the next step for the European strategic State: no longer producing standards alone, but effectively organising the environment in which European companies must apply them.

The issue of SMEs then becomes central. A global company will be able to devote considerable resources to model documentation, risk assessment and compliance procedures. An SME or an innovative start-up will not necessarily be able to do this. The risk would then be that the European normative power, designed to protect citizens and preserve trust in artificial intelligence, would indirectly produce additional market concentration. The issue is therefore particularly delicate: Europe must regulate artificial intelligence without regulating against European innovation.

The Franco-German comparison takes on its full significance here. Germany has an exceptional network of industrial and technological SMEs whose competitiveness depends on their ability to quickly integrate digitalisation and now artificial intelligence. Public support for this transformation is part of a long-standing industrial tradition, in which the State supports the company without necessarily claiming to replace it.¹⁷

France also has a dynamic network of start-ups, mid-caps and innovative SMEs, as well as an ambitious public policy to support artificial intelligence. But at the same time, it retains a more prescriptive administrative tradition. The French challenge is therefore less to create new standards than to succeed in prioritising, simplifying and coordinating existing ones. This difference ultimately refers to a classic question of economic policy, which is particularly important in the France of 2026: how can a state intervene sufficiently to correct market failures without itself creating costs that reduce their effectiveness?

In the industrial economy of the twentieth century, the State could sometimes compensate for administrative complexity through its capacity for investment and planning. In the digital economy of the twenty-first century, the situation is different. Innovation is rapid, technological cycles are short and new companies can appear in a few years. Regulation must therefore itself be able to evolve.

This is perhaps where the most profound difference between the two administrative cultures lies. The French tradition is readily convinced that a difficulty calls for a new rule and willingly stacks it on top of pre-existing rules. The German tradition seems more often to seek to determine, with pragmatism, how an existing rule can be adapted, tested or applied differently depending on the circumstances. The first logic produces visible regulatory power. The second seeks more regulatory efficiency.

This distinction will be decisive for the next phase of European digital sovereignty. The DMA and the AI Act give Europe considerable normative power. But this power will only become a real economic power if it allows European companies to grow, innovate and compete with American and Asian companies. The success of European digital sovereignty will therefore not only be measured by the amount of sanctions imposed on large platforms or the number of regulations adopted. It will also be measured by Europe’s ability to bring out the companies that, tomorrow, will be able to compete with these same platforms, as the Draghi report tried to highlight in 2024.

Conclusion

The digital economy has not removed state sovereignty; it has transformed it. Global platforms have shifted some economic power beyond national borders, but they have simultaneously forced states and the European Union to invent new regulatory instruments. From competition law to the DMA, from cybersecurity to artificial intelligence, the state has gradually moved from the role of belated bystander to that of preventive regulator.

However, a comparison between Germany and France reveals that the Europeanisation of regulations has not yet produced a homogeneity of the digital economies. The Uber case shows this with particular clarity. Germany is more inclined to integrate new technology into the traditional sector, to the point of making the platform an instrument for modernising the taxi. France has built a separate VTC market, then gradually multiplied the measures intended to regulate its professional, social, competitive and administrative dimensions.

This difference does not lead to the conclusion that the German regulations are less demanding. Above all, they appear to be more subtle in their articulation with the economic functioning of the market and more dynamic in their adaptation to the needs of companies, in particular SMEs and the Mittelstand. The French regulations appear, in comparison, to be more abundant and more prescriptive. Its weakness is therefore not necessarily the excess of protection that it pursues, but the risk of an accumulation of standards and implementing administrations whose economic effects are not always sufficiently measured.

This distinction is essential. A large global platform can withstand complex regulations: it has the lawyers, engineers and financial resources necessary to bring it into compliance. An innovative SME cannot always do this. A regulation intended to contain dominant positions can therefore, paradoxically, consolidate these positions by increasing the barriers to entry. The State that regulates without measuring the differentiated cost of the standard thus risks unintentionally protecting the players it precisely wanted to constrain.

This is why digital sovereignty cannot be measured solely by the quantity of rules produced. It is measured by the state’s ability to build a framework that is sufficiently protective to preserve competition, workers and consumers, but simple enough, adaptable and predictable enough to allow companies to take risks, invest and grow.

The AI Act opens a new stage in this respect. Europe is beginning to harmonise the rules that will govern algorithms and artificial intelligence. But this harmonisation will not do away with the States: it will shift their sovereignty towards the application, control, support of companies and the way in which each administration will translate the European standard into economic reality.

The true European digital sovereignty will therefore be that of a Europe capable of regulating without self-regulating against itself. The strategic state of the twenty-first century will not be the one that produces the most standards, but the one that knows how to distinguish between those that protect economic power and those that hinder it. In this respect, the German experience offers France less a model to copy than an essential question: how to move from a culture of regulatory accumulation to a culture of effective regulation?

Annex — Two national architectures of digital regulation: Germany and France

The comparison of the two systems should not be understood as an opposition between a "deregulated" Germany and a "regulated" France. Both states apply the same European law and have a dense set of national rules. The difference lies more in the administrative architecture and the way in which the rule is adapted to economic activity. Germany tends to favour the gradual integration of digital regulation into existing sectoral frameworks, by concentrating expertise and retaining specialised interlocutors; France is making greater use of the simultaneous mobilisation of several administrations and authorities, to which sectoral, social, fiscal and digital obligations are gradually being added.

Notes

1. Regulation (EU) 2022/1925 of the European Parliament and of the Council of 14 September 2022 on contestable and fair markets in the digital sector, OJEU, L 265, 12 Oct. 2022, pp. 1-66; Regulation (EU) 2024/1689 of the European Parliament and of the Council of 13 June 2024 laying down harmonised rules on artificial intelligence, OJEU, L 2024/1689, 12 July 2024, pp. 1-144.

1. François Souty, "European Digital Markets Act, competition policy and sovereignty: geopolitical consequences and strategic impact of the law on the digital economy", Le Diplomate Média, 4 Feb. 2026; id., "Artificial intelligence, normative sovereignty and geopolitics: the fragmentation of global governance between technological powers", Le Diplomate Média, 3 March 2026; id., "Cyberspace, security, sovereignty, technology and global geopolitical rivalries: legal issues, regulation and priorities for the European Union and France", Le Diplomate Média, 11 Feb. 2026; id., "The European Commission and Google (2000-2026): from institutional inaction to the attempt at ex ante regulation of digital platforms", Le Diplomate Média, July 2026. 2026.

1. François Souty, "The European Commission and Google (2000-2026)", supra, in particular the developments devoted to the transition from ex-post control to the use of ex ante regulation of the large platforms; Mario Draghi, The Future of European Competitiveness, Brussels, European Commission, Sept. 2024, spec. developments relating to technological competitiveness, innovation and European regulation.

1. François Souty, "Cyberspace, Security, Sovereignty, Technology and Global Geopolitical Rivalries", supra.

1. Bundesministerium für Wirtschaft und Klimaschutz, "Evaluation zeigt: Förderprogramme für Digitalisierung des Mittelstandes wirken", 6 March 2025; The evaluation notes, inter alia, that the Mittelstand-Digital programmes have improved the degree of digitalisation of SMEs and produced positive effects on their development, while identifying as their main obstacles the lack of financial and time resources, insufficient digital skills and legal uncertainties. In particular, the Digital Jetzt programme is said to have generated, for €134.3 million in aid paid between September 2020 and December 2022, around €447.1 million in increased turnover for the beneficiary companies; European Commission, 2026 Country Report Germany, 3 June 2026, section "Innovation, business environment and productivity".

1. European Commission, 2026 Country Report France, 3 June 2026, including developments relating to the business environment, productivity, innovation and digital transformation.

1. French Competition Authority, Opinion No. 23-A-18 of 29 November 2023 on the competitive functioning of the land passenger transport sector, spec. pp. 88-91 and 448 et seq. for private public passenger transport; The Autorité distinguishes in particular between the marauding market, on which taxis benefit from a legal monopoly, and the prior booking market, on which taxis and VTCs are in competition. It also notes the specific framework for taxi fares and the territorial freedom of the VTC professional card.

1. Personenbeförderungsgesetz (PBefG), § 49, relating in particular to rental cars and the obligation to return to the company’s registered office after the performance of the journey if no new order has been received; see also the provisions resulting from the recent reforms of the system of transport booked by application.

1. Uber Deutschland, "Uber-App soll künftig in ganz Deutschland verfügbar sein: Uber baut dafür Partnerschaft mit Taxibranche weiter aus", 11 June 2024; Uber Deutschland, "Uber ist verlässlicher Partner für Taxifahrer und -unternehmen in Deutschland", 26 June 2025. In June 2024, Uber announced the opening of its platform to taxi companies throughout Germany; the company said that more than 4,000 taxi drivers were already active on the platform. In June 2025, it reported several thousand partner drivers and in particular a partnership with the Stuttgart taxi centre. All data provided by Uber.

1. Uber Deutschland, ‘Zwei Jahre flexible Taxi-Vorab-Preise: Uber zieht positive Bilanz und feiert Meilenstein’, 28 August 2025. Uber states that, since September 2023, pre-determined flexible fares have been tested in nine German cities and that several million taxi rides have been booked through it; the company also states that more than 10 000 taxi drivers were then accessible on its platform. Data provided by Uber and does not constitute, in itself, an independent measure of the evolution of revenues or profitability in the sector.

1. French Competition Authority, Opinion No. 25-A-03 of 21 January 2025 relating to the agreement of 19 December 2023 strengthening the freedom of choice of their rides by VTC drivers using a networking platform, in particular pp. 2-3 and 22-33.

1. Ibid., spec. pp. 24 and 30-33. The Autorité notes in particular that Uber has a particularly significant weight on the French market and that only this operator therefore appears to be in a position to implement the technological device provided for in the agreement; however, it considers that it is not possible, as it stands, to assert that the acquisition of an adequate technological device would constitute an exclusionary factor for competitors. It nevertheless recommends a prior in-depth study of the economic, social, financial and technological consequences before any extension of the agreement.

1. Bundesministerium für Wirtschaft und Klimaschutz, "Evaluation zeigt: Förderprogramme für Digitalisierung des Mittelstandes wirken", 6 March 2025, supra. The evaluation highlights in particular the role of the Mittelstand-Digital centres in supporting SMEs, transferring skills between research and business, and reducing barriers related to the lack of digital skills and legal uncertainty.

1. Regulation (EU) 2024/1689, above, in particular Art. 5 on prohibited AI practices, Art. 6 et seq. on high-risk AI systems, Art. 50 on transparency obligations and Art. 51 to 56 relating, inter alia, to general-purpose AI models.

1. European Commission, AI Act Service Desk, "When does enforcement start?", update 2026. The regulation entered into force on 1 August 2024; the general provisions and prohibitions became applicable on 2 February 2025, the rules on general-purpose AI models and governance on 2 August 2025, while the majority of the rules, including transparency obligations and enforcement powers, apply from 2 August 2026. The rules on high-risk AI systems referred to in Annex III will apply from 2 December 2027 and those on high-risk systems embedded in regulated products from 2 August 2028.

1. For Germany: Bundesnetzagentur, "Bundesnetzagentur übernimmt zentrale Rolle bei der Umsetzung der KI-Verordnung", 29 July 1918. 2026. The German Act on Market Surveillance and Support for Innovation in the Field of Artificial Intelligence (Gesetz zur Marktüberwachung und Innovationsförderung von künstlicher Intelligenz, KI-MIG), which entered into force on the same day, entrusts the Bundesnetzagentur with a central role in the application of Regulation (EU) 2024/1689: it becomes the market surveillance authority for the areas under its competence, the Single National Point of Contact and the central point for the reception of complaints relating to infringements of the Regulation. It also coordinates the Koordinierungs- und Kompetenzzentrum für die KI-Verordnung (KoKI-VO), which is responsible for providing technical expertise to the federal and Länder authorities and coordinating their cooperation. However, the architecture adopted remains hybrid: in sectors already subject to specialised supervision, such as financial services or the media, the existing authorities, in particular the Bundesanstalt für Finanzdienstleistungsaufsicht (BaFin) and the media authorities of the Länder, retain their competences; the Bundesnetzagentur ensures central coordination and intervenes directly in sectors that have not previously been sufficiently regulated, while the Länder can delegate certain competences to it. The system thus combines centralisation of coordination, sectoral specialisation and support for innovation, in particular by means of a regulatory sandbox and a support service aimed in particular at SMEs and start-ups.

As far as France is concerned, the situation is much more complex and is based on a stacking of competences from multiple administrations and agencies: Directorate-General for Enterprise (DGE) and Directorate-General for Competition, Consumer Affairs and Fraud Control (DGCCRF), "The competent authorities for the implementation of the European regulation on artificial intelligence", 9 September 2025. The scheme proposed by the Government is based on a decentralised organisation favouring sector-specific administrative and regulatory authorities. The DGCCRF ensures operational coordination and is the single point of contact within the meaning of Article 70(2) of Regulation (EU) 2024/1689; the DGE ensures strategic coordination and represents France in the European AI Committee. The CNIL, Arcom, the Autorité de Contrôle Prudentiel et de Résolution (ACPR) and various sectoral authorities are involved in their respective fields. The French National Agency for the Security of Information Systems (ANSSI) and the Digital Regulation Expertise Centre (PEReN) provide shared technical expertise. This plan still had to be submitted to Parliament by legislation.

1. Bundesministerium für Wirtschaft und Energie, Mittelstand-Digital programmes and evaluation of support schemes for the digital transformation of SMEs, 2025.

1. European Commission, 2026 Country Report France, supra.

1. François Souty, "The State as a strategist in Europe and France since China’s accession to the WTO (2001-2026): reindustrialisation, competitiveness, economic sovereignty and effectiveness of public action", Le Diplomate Média, 23 July 2026. 2026.

1. François Souty, "Charles de Gaulle, the State as a strategist and the economy (1958-1974): What inspirations for France in 2027?", Le Diplomate Média, 4 Sept. 2026.

#DigitalMarketsAct, #ArtificialIntelligenceAct, #Uber, #France, #Compétition, #Innovation, #Technologie, #Économie, #Concurrence

• ECONOMY – Charles de Gaulle, the Economy and the State as a Strategist  (1958-1974): What inspirations for France in 2027?
• ECONOMY – Strategic State and Industrial Sovereignty of France after 25 years of slowdown : Ideas from Europe, Asia and the United States for Reindustrialization
• ANALYSIS – Ceuta: Trump Denounces a “Catastrophe” and an “Invasion” – 50,000 to 60,000 Migrants, Madrid Under Fire
• ANALYSIS – Trump Warns China of ‘Big Problems’ Over Iran as Xi Summit Nears: Washington Hardens Its Stance
• ANALYSIS – Marco Rubio: The Cuban-American Pivot of Trumpian America Facing an Agonizing Castroism

François Souty

François Souty est Président exécutif du Cabinet LRACG Conseil en stratégies européennes et droit de la concurrence, enseignant à Excelia Business School (La Rochelle-Tours-Cachan), à l’Université Catholique de l’Ouest (Niort) et chargé d’enseignements à la Faculté de Droit de l’Université de Nantes. Auparavant Expert National Détaché auprès de la Commission Européenne (rapporteur antitrust sur les marchés financier de 2018 à 2021 et chargé d’affaires internationales de concurrence à la DG Concurrence de 2021 à 2024), il a été conseiller économique européen pour la politique de la concurrence auprès du gouvernement de Géorgie à Tbilisi en 2017-2018. Longtemps Directeur départemental de la DGCCRF au ministère de l’Économie et des Finances (1982 à 2024), il a été également professeur-associé à l’Université de La Rochelle (1996-2018). Membre des comités d’experts de la concurrence de l’OCDE et de la CNUCED de 1992 à 2018, il a participé aux travaux de l’OMC sur le commerce international et la politique de la concurrence de 1997 à 2004. Un des fondateurs du Cercle Jefferson, du Cercle K2, de la revue Concurrences en 2004, il est auteur d’une douzaine de livres ou rapports internationaux et de plus d’une centaine d’articles académiques en droit et politique de la concurrence et en histoire économique. Il prépare actuellement la 5e édition de «Droit et politique de la concurrence de l’Union Européenne »  chez LGDJ-Montchrestien (coll. Clefs). Il est auteur d’une thèse de doctorat en histoire économique à l’Université de Paris III sur les monopoles des Compagnies des Indes néerlandaises au XVIIIe siècle. François Souty est Officier de l’Ordre National du Mérite.

▼ Lire la biographie complète

Voir tous les articles de cet auteur →