___
An opinion piece in the New York Times last year cited a survey of over 700 top academics and researchers, half of which were of the opinion that there was a “10 percent or greater chance of human extinction (or similarly permanent and severe disempowerment) from future A.I. systems.” This news was covered by other reputable sources including The New Yorker and Vox. The fear mongering about Artificial Intelligent systems (AI), most vividly at display in UK’s AI summit last year, is perhaps only matched by its hype, both of which ascribe sentient, superhuman abilities to systems which are essentially statistical models running on very large quantities of data. These over-inflated fears turn our much needed attention away from the real-life dangers and risks of AI, which are numerous and far more prosaic. As we write this, machine learning algorithms (ML) intermediate our use of everyday services which determine how we spend our time, what content we read, view and consume, what opportunities we pursue and perhaps, what we think. Despite their ubiquity in diverse aspects of life, the common denominator of AI/ML remains their opacity.
Machine learning algorithms (ML) intermediate our use of everyday services which determine how we spend our time, what content we read, view and consume, what opportunities we pursue and perhaps, what we think. Despite their ubiquity in diverse aspects of life, the common denominator of AI/ML remains their opacity.
~
Over the course of the next three essays, I will demonstrate that the domain of administrative law is uniquely suited to guide the defining of thresholds for meaningful algorithmic transparency, both in public and private law. In this first essay, I will provide a brief overview of what makes administrative law suitable. Below, I will highlight the principles of administrative law which render themselves appropriate for application in the domain of algorithmic transparency and accountability and speculate how they may be applied to govern AI.
What is administrative law?
Administrative law is concerned with constituting, limiting, and holding to account public administration and law includes a plurality of means for doing this. The public administrative organism is a large and varied authority “to which power has been delegated from a primary lawmaker,” [1] even more in a privatised state. The administrative setup includes a range of institutions with diverse information, expertise, discretion, and deliberation processes. As AIs becomes more pervasive across sectors and institutions, if not an obvious, certainly a useful parallel can be drawn between the delegation of decision-making from humans to machines and the delegation of power from identifiable and elected lawmakers to recondite and impenetrable administrative institutions.
Given the labyrinthine nature of its subject, the primary business and as Elizabeth Fisher puts it, the “central obsession” of administrative law have been accountability. I will demonstrate how the intricate discipline of administrative law and the principles which have evolved in common law to hold its Daedalian institutions to account, are instructive in how we must make AIs, whether used by the public bodies or private parties, more scrutable.
The domain of administrative law deals with secrecy and disclosure and offers several guidelines on why, when, and how to disentangle the dramatically concealed actions of public bodies. The reasons behind secrecy range from the deliberate guarding of secrets to the complex and technical nature of public administration, from institutional design to evade accountability to less than meaningful transparency measures where access to information does not translate into understandability. For any researcher or practitioner of algorithmic transparency, these obstacles to meaningful transparency in the domain of administrative law will sound eerily similar.
When we seek more transparency, what we seek can vary immensely based on the context and circumstances of the institutions and stakeholders involved, the nature of the information we are dealing with, how the information is being delivered, who the recipient of the information is and the consequences of disclosure or inversely secrecy. In this section, I will present four essential factors that determine the design of any transparency initiative, as drawn from public administration but having an equally relevant role for algorithmic transparency.
Why is transparency sought?
When introspecting on the rationale behind transparency, it is useful to refer to MacIntyre's distinction between primary virtues and secondary virtues. The first relate directly to the goals which we pursue, while the second are concerned with the way in which we go about them. As I stated earlier, it may be perhaps most accurate to consider transparency as occupying a space somewhere between the primary and second virtues. [2]
As a secondary virtue, transparency is key for the public to achieve greater autonomy and facilitate greater accountability of institutions. However, relegating it entirely to a secondary virtue may translate into placing a burden in each instance to demonstrate that transparency will lead to some policy goals. As we have discussed earlier, it is reasonable to presume that more publicly available information is preferable to less publicly available information, all else being equal.
What needs to be transparent
It is useful to think of transparency as a sliding scale with full secrecy at one end and complete disclosure at the other. The design of the transparency initiative is often about making choices about what needs to be disclosed. It is almost never that complete disclosure is warranted or even desirable. The laziest approach to deciding what to disclose is to reveal a resource that already exists. Therefore, most of the freedom of information legislation will require access to documents. In many cases, information is created primarily for the purpose of disclosure, for example, quarterly and annual financial accounts are compliance and reporting requirements in several contexts. The duty to give reasons for administrative and judicial bodies requires the explicit recording of reasons and the creation of an essential resource.
There are also cases where the demand for disclosure precedes the creation of the documents. These may include several kinds of analysis, risk assessments, performance indicators etc. What is being made visible by a transparency initiative is a resource that those within an institution are drawing on for their power. Such resources are often frameworks (legal, organisational, technical), the information or expertise, or the normative values on which the decisions are based. Often it may be required to create new resources to achieve transparency.
As we can see, what needs to be made transparent varies greatly, and its most useful determinant is often what information can be furnished, without imposing more than necessary costs of production, such that the intended recipient of the information can achieve the purposes of the transparency initiative.
When should it be made transparent
There are two facets to the temporal nature of transparency. The first is the trigger for transparency — when does the disclosure or its requirement kick in. The second is the time period for which it remains transparent and then reverts to secrecy.
The trigger for transparency, as Fishcer characterises it, again varies greatly. There are transparency mechanisms which are inbuilt into the design as periodic. The creation and disclosure of documents does not rely on an external trigger but rather a pre-defined passage of time or event. These may include quarterly publication of accounts or publication of minutes after each meeting.
In some cases, the transparency initiatives are designed to precede the decision-making process. Therefore, impact assessment reports must be prepared before they may inform the decisions. The notice and comment form of policymaking where drafts are made public for comments is another example where the transparency measure is designed to aid decision-making.
There are also external triggers, most notably in freedom of information legislations where the request for information leads to disclosure though, in most cases, the creation of documents precedes the trigger.
The trigger could also arise because of a dispute where documentation needs to be mandatorily shared between parties as a part of the evidence.
In yet other cases, the transparency mechanism kicks in only after a sufficient time has passed after the event in question. The Necessity and Proportionality Principles ask for notice to be provided to individuals subjected to surveillance after the fact of surveillance. This is a case of a decision and action that, by its very nature, require secrecy. [3] However, transparency requires that such secrecy cannot be sanctioned for eternity, and after the act and the passage of a reasonable period, it must be disclosed. A similar logic also applies to the declassification of national security documents years later.
The second facet of the temporal aspect of transparency sees less range of application, now that digital publication has reduced the costs of transparency. When dealing with transparency of administrative bodies, there is no clear argument for public interest in access to the documents extinguishing after a period. However, practical considerations of costs to maintain the public records could be responsible for their withdrawal or deletion.
Who is the recipient of transparency?
The next facet of transparency initiatives relates to the intended recipient of the information. When we think of transparency, the first instinct is to imagine all citizens or consumers as its audience. In the case of freedom of information legislations, this is true. In several cases, the primary recipients of transparency are the regulated entities and no one else. [4]
In some cases, what is being made visible, by its very nature, require expertise and training to consume, such as financial reports, or technical assessments. In such case, even though most people cannot be expected to meaningfully engage with the documentation, they rely on other experts to educate the public about it.
What is implicit is each case, is that those creating the disclosure must have an idea of the intended recipient, and accordingly keeping in mind what the recipient reasonable requires.
Conclusion
The rich tradition of administrative law and the complex body of public administration that it governs provides us with very useful questions as we approach algorithmic transparency. As much as is made of the inherent opacity of AI, we often tend to approach its transparency obligations with a need to reinvent the wheel. On the contrary, we need to more stringently rely on the ways to approach transparency that have served us well. In the next two essays, I will throw light on clear principles of transparency in administrative law that can be applied directly to achieve algorithmic accountability, both in public and private sector use of AI.
[1] For a more detailed analysis of the nature of administrative power, please see Athanasios Psygkas. 2021. Accountability in Peter Cane, Herwig Ch Hofmann, Eric C Ip, and Peter L Lindseth. 2021. The Oxford Handbook Of Comparative Administrative Law. Oxford University Press, Oxford.
[2] In The Rise of the Right to Know: Politics and the Culture of Transparency, Michael Schudson elaborates on this midway space that transparency occupies.
[3] Principle 8 of the Necessity and Proportionality Principles provide an excellent and comprehensive example of post-facto exercise of transparency.
[4] A good example is Article 5 of the EU Regulation on the negotiation and implementation of air service agreements between Member States and third countries.