Innovation Law and Regulation 101: AI as Civil Servants

Foreword

Artificial Intelligence systems are set to be the next revolution, forever changing humans’ lives. This new phenomenon and its many effects will cause great changes in our society, which is why regulating is the first step toward ethical development. In fact, unregulated use of these technologies could give rise to negative consequences such as discriminatory uses and disregard for privacy rights. The challenges brought by the use of Artificial Intelligence urge legislators and experts to protect citizens and consumers as regulating becomes a priority if humans wish to protect themselves from unethical and abusive conduct. This series explores the topic of new technologies, such as artificial intelligence systems and their possible regulations through legal tools. In order to do so, we will start with an explanation of the rise of new technologies and delve into the complicated question of whether machines can be considered intelligent. Subsequently, the interplay between Artificial Intelligence and different branches of law will be analyzed. The first chapter of this series of articles explored the possibility of granting AI systems with legal personality and the main legislative steps taken in the EU towards that direction. Moving into the realm of civil law the second chapter considered the current debate on the responsibility regime concerning the use and production of AI. The third chapter will discuss the influence that AI plays on contract law and the stipulation of smart contracts. The use of AI in criminal law and the administration of justice was examined in the previous instalment with a focus on both the positive and negative implications of their use. The present chapter will be dedicated to the use of Artificial Intelligence by public sector bodies and how new technologies could improve the field of administrative law. Finally, the complicated relationship between data protection, privacy, and AI will be discussed in the following and last article of this 101 series in light of the EU General Data Protection Regulation.

The 101 series is divided into six articles:

1. Innovation Law and Regulation 101: Recognizing Silicon Minds

2. Innovation Law and Regulation 101: AI on Trial, Blaming the Byte

3. Innovation Law and Regulation 101: Navigating Smart Contracts

4. Innovation Law and Regulation 101: Silicon Justice

5. Innovation Law and Regulation 101: AI as Civil Servants

6. Innovation Law and Regulation 101: Defending Data from Silicon Eyes

Innovation Law and Regulation 101: AI as Civil Servants

The advent of Artificial Intelligence has transformed many aspects of how human lives are carried out. Law is no different as its many fields have been affected by the use of Artificial Intelligence systems challenging the traditional methods that have been employed until now. Administrative law is an extraordinary example of how new technologies could improve efficiency and overall performance for the benefit of citizens. However, the integration between Artificial Intelligence and administrative law also raises crucial questions and challenges in terms of the flaws still unsolved in the use of new technologies. This article will explore the intricate relationship between administrative law and AI to illustrate the advantages and disadvantages of this new combination. The first part will be dedicated to an explanation of what administrative law is and the principles governing this field of law. Subsequently, the advantages that the use of Artificial Intelligence systems might lead to will be analysed, particularly their efficiency, impartiality, and improvement toward more legal certainty. The following paragraphs will delve into the still unsolved issues of Artificial Intelligence, with emphasis added on the concerns about its possible bias, lack of transparency, and the raising concerns for the impossibility of ensuring the full application of the rule of law and an efficient liability regime. A case brought before an Italian administrative court will also be illustrated with the aim of providing a practical example of the negative implications of AI tools. The final part will be dedicated to conclusions.

Administrative Law: From the Origins to the New Principles

Administrative law is a branch of public law that governs the activities of public sector bodies of the State (Harlow & Rawlings, 2019). This is a relatively young field of law as its formal recognition goes back to the end of the nineteenth century when industrial revolutions led to a growing number of welfare tasks being carried out by the State (Harlow & Rawlings, 2019). The phenomenon was followed by the spread of a new legal theory in the United States known as functionalism, which dealt with the fact that the State and its functions had begun to change, and among its new tasks, it now had to implement public services through clear and detailed administrative processes (Harlow & Rawlings, 2019). As a consequence, the State now had to focus on ensuring a good administration for the benefit of all citizens. In Europe, the beginning of administrative law emerged more clearly in France and England where, by the beginning of the nineteenth century, bodies made up of public officers started to be established to perform specific tasks of public interest (Rose-Ackerman, Lindseth & Emerson, 2017). The new rules established in France gave special powers to public officers in light of the importance of their roles, which were carried out in the interest of society as a whole (Tocqueville, 1988). These public bodies were allowed, in fact, to make unilateral decisions that could result in restrictive measures against private citizens, such as orders of land expropriations (Rose-Ackerman, Lindseth & Emerson, 2017). Moreover, the disputes between public and private parties were not solved by the ordinary justice circuit but rather by a special judge who was the only one vested with the jurisdiction over administrative law controversies (Rose-Ackerman, Lindseth & Emerson, 2017).

Figure 1: Administrative Tribunal of France (Nguyen, 2006).

The rather authoritarian framework in which administrative law developed was however doomed to find its end as more liberal instances began to be voiced by citizens (Giannini, 1993). Public administrations now had to balance their special powers with the interests of private citizens, which were given more importance as the years went by. In fact, private interests were finally starting to find formal recognition and protection through the spread of liberal constitutions in Europe (Rose-Ackerman, Lindseth & Emerson, 2017). The so-called welfare State was what functionalists had been theorizing all along in the United States during the nineteenth century: a democratic State ensuring efficient services for its citizens. This was reflected in the application and development of new principles concerning the functioning of public bodies. Unilateral acts by public bodies were now subject to a principle of compatibility with public interest and balance between opposing private interests (Plessix, 2003). Furthermore, in many European countries, administrative law acts could be reviewed and challenged by courts and deemed illegal for excess use of power (Rose-Ackerman, Lindseth & Emerson, 2017). The process culminated in the formal recognition of multiple principles provided by both the European Union as a whole and single States as well. For the purpose of this essay, it is fundamental to illustrate the most relevant ones.

The principle of proportionality is perhaps one of the most important principles governing the work carried out by public institutions. Article 5(4) of the Treaty on the European Union provides that the "content and form of Union action shall not exceed what is necessary to achieve the objectives of this Treaty" (Treaty on the European Union, 2009). The principle was also subject to extensive interpretation by the European Court of Justice, and entails that public bodies should do only what is strictly necessary to achieve their objectives legitimately pursued (Barnard & Peers, 2014). Moreover, when there is a choice to be taken between several appropriate measures the public bodies should take the least onerous one (Barnard & Peers, 2014). Finally, the disadvantages that might be caused must not be disproportionate to the aims pursued. The illustrated principle clearly deviates from the notion that public bodies can freely and discretionally use their special powers and provide protection for the interests of private parties.

Figure 2: The signing of the EU Treaty of Lisbon (Wikipedia, 2007).

The principle of transparency, provided by Article 15 of the Treaty on the Functioning of the European Union, requires that all bodies are transparent and provide copies of their decisions as well as access to the documents that might have been used (Barnard & Peers, 2014). The principle of transparency entails that public bodies shall be open and communicative with the parties involved and do so in a timely manner (Jashari & Pepaj, 2018). Scholars also interpret this principle as key to a democratic society and a pillar in a State that wants to promote participation and involvement of citizens (Curtin & Meijer, 2006). Thus, the relationship between transparency and legitimacy is fundamental, as only a transparent institution can and will be trusted to carry out its tasks in the interest of citizens. Article 41 of the Charter of Fundamental Rights is expressively titled “Right to Good Administration” and provides that individuals should have access to files relating to their affairs. Furthermore, Article 42 of the Charter of Fundamental Rights provides that any citizen has a right to access the documents of public institutions.

The principle of legal certainty is a shared value among most democratic legal frameworks. In fact, it is qualified as a multi-faceted principle and underpinning of any legal system (Portuese, Gough & Tanega, 2017). It is sometimes referred to as a right to legitimate expectations as it is about the fact that the law, and the many acts issued by public bodies, should be clear and predictable (Caranta, 2023). It also entails that the law should not be retroactive as it would violate the notion that the law shall be known to individuals and its application foreseeable (Portuese, Gough & Tanega, 2017). Another corollary of this principle is that the application and interpretation of the law should be consistent throughout time because the application shall derive from clear provisions subject to as little discretion as possible. It is, therefore, clear that citizens can be fully protected and shielded against arbitrary decisions by public bodies when the principle of legal certainty is respected (Grinbau, 2014).

Figure 3: Preamble of the EU Charter of Fundamental Rights (Trounce, 2008).

The rule of law is another key principle governing administrative law. Its fundamental meaning is that all public institutions related to the State shall act in accordance with the existing law (Grinbau, 2014). In fact, the law is conceived as a general and abstract norm capable of regulating in an impartial manner, and that is why it is a central pillar of liberal democracies (Dallmayer, 1992). John Locke, the renowned English philosopher, defined laws as constant-based rules equal for all, and as such, it is the law that can ensure that powers are not used arbitrarily but according to rules (Locke, 1988). Article 97 of the Italian Constitution is a perfect example of how national frameworks in Europe have implemented said principle: it provides that public offices shall be regulated by the law to ensure efficiency and impartiality. Furthermore, the rule of law entails that administrative acts shall be the result and last step of an administrative procedure regulated by the law, and the law alone can give public officials the power to use their discretion when needed (Rose-Ackerman, Lindseth & Emerson, 2017).

It is impossible not to underline the importance of the general principle of good administration laid out by Article 41 of the European Charter of Fundamental Rights. The provision, also recalled in the previous paragraphs, entails that administrative acts must be followed by detailed reasoning (Barnard & Peers, 2014). Article 296 of the Treaty on the Functioning of the European Union provides for the same obligation for all European Union institutions and describes what a statement of reasons should contain. It lists as fundamental elements of the statement of reasons the motivations, explained clearly, as to why the decision was taken and the indication of which body took it. That is essential to enable individuals the possibility to seek redress and judicial review of administrative acts, leading to another fundamental pillar of a good administration: accountability of public bodies.

Figure 4: English Philosopher John Locke (Greenhill, n.d.).

The principle of accountability is a synonym for the right of every citizen to receive effective remedy in a competent court (Barnard & Peers, 2014). Because the powers of public bodies are not unlimited, it follows without saying that public officials should be held accountable for their actions, especially those damaging the rights and interests of private parties. This is also linked to the need for public officials to ensure that all their decisions and discretionary acts are followed by a statement of reasons. In fact, such a statement ensures that the motivations behind decisions are easily traceable to the responsible party. Moreover, private parties have the right to seek judicial review of any administrative act issued by public bodies (Barnard and Peers, 2014). During eventual proceedings before an impartial and competent court, all principles regarding the right to defense and fair trial are also to be ensured. Courts also have the power to annul administrative acts when they violate one of the principles governing administrative law or when excessive in the use of power. The latter notion, conceived in German law, entails that administrative acts are illegitimate when not issued in accordance with procedural law regulating the administrative process or when issued by "incompetent parties" (Petoft, 2020). Incompetent parties are public officials who were not invested by the law with the power to exercise their role and emanate acts, according to the principle of the rule of law (Carroll, 2009).

The principle of impartiality and fairness does not find application in administrative law alone: it is a general principle recognized as a pillar of democratic states and, therefore, also finds its way into the realm of administrative law. Article 14 of the European Convention of Human Rights provides that all individuals shall be treated equally, and it prohibits any act of discrimination. This entails that public officials cannot take discriminatory or partial actions toward any individual and that situations that are objectively the same shall be treated equally (Rossi and Casolari, 2017). All administrative acts shall be the result of an impartial exercise of power (Petoft, 2020).

Figure 5: Ukranian stamp commemorating the 60th anniversary of the European Convention on Human Rights (Wikipedia, 2010).

In order to ensure impartiality and transparency, the principle of public participation is also promoted in many legal frameworks (Petoft, 2020). The attempt to make citizens more involved in the decision-making process can, in fact, lead to a better-balanced administrative act where all interests at play are carefully taken into consideration. Moreover, it might guarantee higher standards of transparency, which could favorably contribute to legitimizing the power attributed to public bodies (Rohl, 1993). The Italian law on the administrative process specifically protects said principle by providing that any individual can intervene during the decision-making process that precedes the issuing of the final administrative act (Article 9, Law 241/1990).

Through the principle of efficiency, it is ensured that administrative processes are carried out efficiently in order to guarantee fast and clear actions by public officials (Figueiredo, 2009). The principle of efficiency is also strictly related to the concept of economic action (Manzoor, 2014). Action should be efficient, and it can be so only when made with the most effective yet least expensive tools available. Furthermore, efficiency is also related to good time management, entailing that an administrative process should not take more time than necessary (Finck, 2019). This is fundamental to ensure that citizens can benefit from a good administration as provided by Article 41 of the European Charter of Human Rights. Article 3-bis of the Italian law on the administrative process interprets the principle of efficiency as the principle of telematics. In fact, the provisions state that public bodies shall use telematic and technological tools to promote efficiency. This leads to the question of whether Artificial Intelligence can find full legitimization in the realm of administrative law and not violate the above-mentioned principles governing the administrative process.

Figure 6: Interplay between AI and Administrative law (UNDP, n.d.).

Artificial Intelligence and Administrative Law: a Match Made in Heaven?

Artificial Intelligence systems have the extraordinary advantage of being able to emulate human thinking without suffering from human flaws (Korteling et al., 2021). While humans have limited cognitive skills, computers' capacities have potentially no limits (Korteling et al., 2020). For once, humans cannot perform demanding tasks simultaneously, and human brains are capable of only focusing on one executive cognitive action at once (Rubinstein, Meyer & Evans, 2001). Furthermore, human memory is limited to both the time it takes to be processed and the time it can be stored in the future as acquired information is doomed to be forgotten over time (Wingfield & Byrnes, 1981). Thus, it is likely that the use of Artificial Intelligence systems in the field of administrative law might lead to positive consequences with benefits for citizens.

New technologies, especially "intelligent" ones, have unprecedented levels of efficiency (Korteling et al., 2020). Their capability to scan and analyze enormous amounts of data might prove essential in particular steps of the administrative process, such as the inspection step (Hermstrüwer, 2020). In fact, most, if not all, administrative processes begin with an inspection of the context and situation to gather enough information to subsequently take action through an administrative act (Hermstrüwer, 2020). During this time, efficiency is essential in guaranteeing that all details are taken into consideration, and only this way correct decisions will be made in the future. In fact, correct decisions are likely to be the result of a thorough consideration of all elements pertaining to the situation object of the process. As humans, public officials are subject to stress, fatigue, and generally limited cognitive capabilities, whereas Artificial Intelligence systems are not and have proven themselves able to gather more information and analyze it in less time (Noy & Zhang, 2023). This could lead to an enhanced application in real life of the principles of efficiency and good administration. In fact, to ensure that those principles and the laws providing for them do not remain deadlocked, it is essential that legislators and public bodies find solutions to inefficient administrative processes, and the use of Artificial Intelligence might represent a step forward in that direction.

Figure 7: The many fields of A.I. (Intellipaat, 2023).

Algorithmic tools run by Artificial Intelligence technologies could be also employed during the decision-making step. In fact, after an inspection, public bodies are called to make decisions, and they do so through the issuing of a final administrative act. The moment is even more relevant in discretionary processes, which give more power to public bodies who are legitimized to use their own discretion as a criterion to make a final decision. Scholars already suggest that machine learning systems might be employed to first analyze the context by being fed with relevant information and then make a decision (Coglianese & Lehr, 2019). For instance, the algorithm might be used to analyze a certain situation pertaining to tax evasion and then decide whether the situation amounts to a full or partial violation and the related sanctions that might be applied (Coglianese & Lehr, 2019). In the near future scholars are already envisioning that the principles of impartiality, fairness, and legal certainty would be better protected this way (Coglianese & Lehr, 2019). In fact, an algorithm makes a decision based solely on its data training and the information of the particular context examined. There is no left space for discretionary elements being added by humans who might be partial to a certain solution rather than another. Citizens could rely more on administrative law while being more certain, and therefore trustful, that an impartial decision will be made. Decisions would also be more consistent ensuring that the principle of legality finds full application.

Active participation is another principle that might be better safeguarded if Artificial Intelligence tools were to be used by public officials. Since administrative processes and acts should be the result of a fair balance between different interests, new technologies could be employed to give the chance to all citizens to have a say in the administrative process. In fact, technologies could be more accessible methods of participation for those people living in rural areas or with motor disabilities. Only a society that does not discriminate against its own citizens can be considered a society in which the principles of equality and active participation are fully understood and applied. The opinions of many could be easily registered with algorithmic tools and analysed quickly and efficiently building a statistical picture of what citizens want and need (Coglianese, 2021).

Figure 8: AI system (European Business Review, 2022).

Artificial Intelligence and Administrative Law: a Doomed Relationship?

Despite the promising results that await in the future, there are also challenges and issues related to the use of Artificial Intelligence tools, which need urgent and efficient solutions. Technology, with all its extraordinary features, lacks one key element: human intelligence. As such, it is unlikely that citizens and legislators will accept its use in the realm of administrative law before it is completely certain that problems will not arise and damage the lives of people. In the following paragraphs, the main challenges in the relationship between administrative law and Artificial Intelligence will be discussed.

First of all, it is essential to consider carefully the principle of transparency, a pillar of a good administration. If decisions need to be transparent, it is unlikely that machine learning, at the moment, can offer a compelling argument for the use of Artificial Intelligence in administrative processes. In fact, algorithmic systems are considered black boxes in which decisions are made in an obscure manner (Coglianese, 2021). Algorithms have a certain degree of autonomy which makes it impossible at the moment to fully understand the process used to make a certain decision and, simply put, why an element A leads to B. The black box phenomenon also entails that decisions are unlikely to be motivated, as the principle of good administration provides for. The principle of transparency provides that the basis for a good administration is the issue of motivated decisions, which explain the exact thought process behind every decision made by a public sector body. The Italian Administrative Tribunal of the Lazio region argued that the use of algorithms in the decision-making process is not possible unless it is ensured that the algorithm is fully understandable by all parties involved (TAR Lazio, Decision n. 12026/2016). In the proceedings, the parties argued that their right to a motivated decision, the corollary of the principle of good administration, and to a transparent administrative process had been violated by the Ministry of Education because it had taken a decision pertaining to the allocation of teachers to regional districts through the use of an algorithm. As a result, the final decision was fully taken by a machine learning system with no possibility for the parties to have access to it nor to receive motivation as to why the particular allocations had been made. For this reason, the Italian judges, in favor of the arguments brought by the parties, declared that the administrative act was invalid. This sheds light on an issue that the legislator must solve in order to guarantee the protection of individuals' rights against the power of administrative bodies.

Figure 9: TAR Lazio (Giustizia Amministrativa, n.d.).

The principle of accountability also might be endangered by the use of Artificial Intelligence systems. In fact, whenever a decision is made employing an algorithm, it is difficult to link the decision to an individual subject to duties and responsibilities. As of now, Artificial Intelligence technologies are not yet recognized as people in a judicial sense and thus are not responsible for the damages caused to other parties (García-Micó, 2020). This entails that if a citizen suffers damages caused by an administrative act and it is not possible to understand which public official was responsible for the use of the algorithm, the citizen might not receive proper judicial remedy. The whole public body could be sued by the citizen, but this might make it easier for the public body to argue for lack of implication in the matter, as administrative liability is usually linked to a sole public official responsible for the administrative act. The fear that public bodies might hide behind algorithms is relevant and needs urgent regulation before opening the door for Artificial Intelligence tools in the realm of administrative law.

Algorithms are, as mentioned above, able to make independent decisions limiting the use of human discretion. However, it is important to underline that all algorithms undergo a process of training using existent data. The problem is that the training data is made up of decisions made by public officials, which are likely to contain varied degrees of bias. In fact, it is a human characteristic to make biased decisions, and the use of algorithms might actually make the situation worse by spreading prejudices even more by making decisions based on statistics (Sunstein, 2019). It is possible that enhanced technologies and scientific progress combined will result in more transparent algorithms, which would allow experts to detect impartial decisions and eliminate them while also training the program to disregard biased data (Johndrow & Lum, 2019). Nevertheless, before using these systems, it is necessary that it is ensured that unfair decisions will be reduced thanks to technology rather than worsened.

Figure 10: Algoritms can be biased (Animashaun, 2020).

Finally, the principle of the rule of law must be considered to fully understand the resistance against Artificial Intelligence in the field of administrative law. As mentioned in the first part of the article, administrative law is governed by the principle of law, and as such all administrative processes and acts, it must comply with the law. For this reason, it is the law and the law only, as crafted by the legislators, that can allow the use of certain methods or allow for the exercise of discretionary powers. Many scholars argue that as long as the law does not provide for the possibility of using Artificial Intelligence systems in the administrative process, their use must be considered in violation of the existing law (Ruffolo, 2020). Given that the principle of the rule of law translates to a general right of individuals to be protected by administrative discretionary power, it is essential that this principle is treated with the utmost attention. For this reason, the first desirable step, even before the solutions to the above-mentioned issues, is a legislative intervention allowing for the use of these new technologies in the realm of administrative law.

Conclusions

In conclusion, the current debate on the possible use of Artificial Intelligence in administrative law is divided into positions in favor and positions against. Favorable positions are based on the characteristics of these new technologies and the benefits that might be experienced by citizens with regard to administrative processes and acts. In fact, Artificial Intelligence tools are expected to lead to more efficient processes and consistent and impartial decisions. Furthermore, it is likely that technologies will enhance the active participation of individuals. However, it is also important to note that some issues and challenges are still unsolved. Particularly, the principles of transparency and good administration might suffer from the use of algorithms. Given that algorithms are considered opaque, it is likely that courts will annul administrative acts that contain decisions taken by automated means, as happened in Lazio, Italy. Moreover, the principle of the rule of law makes it impossible at the moment to envision a future in which Artificial Intelligence systems can be employed without violating the existing law, which does not allow, or even consider, their use. It is essential that action is taken by legislative bodies that have the power to finally allow the presence of new technologies, but not before scientific progress is made to ensure that citizens will benefit from their use rather than suffer from unsolved issues.

Bibliographical References

Ackerman, S., Lindseth, L. P. & Emerson, B. (2017). Comparative Administrative Law. Edward Elgar Publishing.

Barnard, C. & Peers, S. (2014). European Union Law. Oxford University Press.

Caranta, R. (2023). Administrative Proceedings in Italy. In Felix Uhlman (ed.)., Codification of Administrative Law. Hart Publishing.

https://library.oapen.org/viewer/web/viewer.html?file=/bitstream/handle/20.500.12657/74778/9781509954940.pdf?sequence=1&isAllowed=y

Carroll, A. (2009). Constitutional and Administrative Law. Longman.

Coglianese, C. (2021). Administrative Law in the Automated State. American Academy of Arts & Sciences, 150(3), 104-120. https://doi.org/10.1162/DAED_a_01862

Curtin, D. & Meijer, A. (2006). Does Transparency Strengthen Legitimacy? A Critical Analysis of European Union Policy Documents. Information Polity, 11(2), 109-122. https://ssrn.com/abstract=1434862

Dallmayer, F. (1992). Hermeneutics and the Rule of Law in G. Leyh (ed.). Legal Hermeneutics. History, Theory and Practice. University of California Press.

Figueiredo, J. (2009, 26-27 March). Efficiency And Legality In The Performance Of The Public Administration. Conference on Public Administration Reform and European Integration Budva, Montenegro. https://www.sigmaweb.org/publications/JFigueiredo_EfficientPA_Eng_July10.pdf

García-Micó, T.G. (2021). Electronic Personhood: A Tertium Genus for Smart Autonomous Surgical Robots?. In Ebers, M., Cantero Gamito, M. (eds). Algorithmic Governance and Governance of Algorithms. Data Science, Machine Intelligence, and Law. Springer.

Gribnau, H. (2014). Legal Certainty: A Matter of Principle. Tilburg Law School, 12, 69-93. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2447386

Harlow, C. & Rawlings, R. (2020). Proceduralism and Automation: Challenges to the Values of Administrative Law in E. Fisher (ed.). The Foundations and Future of Public Law: Essays in Honour of Paul Craig. Oxford Academic Press. https://academic.oup.com/book/36678/chapter-abstract/321728035?redirectedFrom=fulltext

Hermstrüwer, Y. (2020). Artificial Intelligence and Administrative Decisions Under Uncertainty. In T. Wischmeyer, T, Rademacher (eds.). Regulating Artificial Intelligence. Springer, Cham. https://doi.org/10.1007/978-3-030-32361-5_9

Jashari, M. & Pepaj, I. (2018). The Role of the Principle of Transparency and Accountability in Public Administration. Acta Universitatis Danubius, 10(1), 60-69.

https://core.ac.uk/download/pdf/229465497.pdf

Johndrow, E. J. & Lum, K. (2019). An Algorithm for Removing Sensitive Information: Application to Race-Independent Recidivism Prediction. The Annals of Applied Statistics, 13(1), 189-220. https://arxiv.org/abs/1703.04957

Manzoor, A. (2014). A Look at Efficiency in Public Administration: Past and Future. SAGE Open, 4(4), 1-5. https://doi.org/10.1177/2158244014564936

Noy, S. & Zhang, W. (2023). Experimental evidence on the productivity effects of generative artificial intelligence. Science, 381, 187-192. https://www.science.org/doi/10.1126/science.adh2586

Plessix, B. (2003). L'utilisation du droit civil dans l'élaboration du droit administratif. Éditions Panthéon Assas.

Portuese, A., Gough, O. & Tanega, J. (2017). The principle of legal certainty as a principle of economic efficiency. Eur J Law Econ, 44, 131–156. https://doi.org/10.1007/s10657-014-9435-2

Rossi, L. S. & Casolari, F. (2017). The Principle of Equality in EU Law. Springer Charm. https://doi.org/10.1007/978-3-319-66137-7

Rubinstein, J. S., Meyer, D. E. & Evans, J. E. (2001). Executive control of cognitive processes in task switching. J. Exp. Psychol. Hum. Percept Perform. 27, 763–797. doi:10.1037

Ruffolo, U. (2020). XXI Lezioni di Diritto dell’Intelligenza Artificiale. Giappichelli.

Sunstein, C.R. (2019). Algorithms, Correcting Biases. Social Research: An International Quarterly 86(2), 499-511. doi:10.1353/sor.2019.0024

TAR Lazio. (2016). Decision n. 12026.

Tocqueville, A. (1988). Democracy in America. HarperCollins.

Wingfield, A. & Byrnes, D. (1981). The psychology of human memory. Academic Press.

Visual Sources