Transparency is the political instrument that APIs make possible and that powerful interests resist, and it’s one of the threads I’ve pulled on hardest throughout my work. APIs can make systems transparent — exposing how they work, who can access them, what they cost, and what they do with data — in ways that were never possible before. And precisely because transparency threatens the powerful, the fight over API transparency is genuinely political. Transparency is accountability’s prerequisite: you can’t hold a system accountable if you can’t see inside it, and APIs are uniquely capable of opening systems to scrutiny. The politics of transparency runs through everything from government open data to algorithmic accountability to platform disclosure of who has access to our information. When I’ve argued for transparency through APIs, I’ve been arguing for a redistribution of the visibility that power depends on keeping to itself.
The transparency report is the building block I championed most concretely, and it’s a model worth understanding. I wrote in 2013 about making transparency reports standard operating procedure and about the API transparency report as an essential building block — drawing on Google’s practice of publishing reports about government requests for data. The transparency report is a disclosure mechanism: a regular, public accounting of who asked for what, what was handed over, and what the platform did. I argued that this should be standard for all API providers, and even that transparency report data should be machine-readable and available via API — transparency about transparency. The transparency report matters because it makes visible the otherwise-hidden flows of data and the otherwise-invisible exercise of government and corporate power. It’s a small but real mechanism for holding powerful systems accountable through disclosure.
Algorithmic transparency is where I pushed the transparency argument furthest, because it’s where the stakes are highest. I wrote repeatedly — keeping a window open into how power flows within algorithms using APIs, and most pointedly in 2016 that if an algorithm impacts our life, it should be opened up with an API for auditing. The argument is that algorithms increasingly govern consequential decisions — what we see, what we’re offered, how we’re treated, whether we’re policed — and that these algorithms operate as black boxes immune to scrutiny. APIs offer a mechanism to crack open those black boxes, to let researchers, journalists, and regulators audit how the algorithms actually behave. This is transparency as accountability for the most powerful and least visible systems in modern life. The resistance to it is intense, precisely because algorithmic transparency would constrain the power that algorithmic opacity protects. The fight over whether the algorithms that shape our lives will be transparent enough to audit is one of the defining political fights of the API age.
The data-access transparency dimension is where transparency intersects most directly with privacy and surveillance. I wrote in 2017 about transparency around every company that has access to our social data via an API — because in the wake of the data-sharing scandals, the question of who can see our data became urgent, and transparency about that access is a precondition for any accountability. Cambridge Analytica was, at its root, a failure of transparency: nobody knew who had access to the data or what they were doing with it. The transparency in police access to social platforms that I wrote about in 2016 — the Geofeedia surveillance case — is the same issue in the law-enforcement context: who is accessing social data, for what purposes, and is anyone watching. Transparency about data access is how the people whose data flows through these systems can know what’s happening to it, which is the precondition for having any say over it. The absence of that transparency is itself a political choice that serves the interests of those doing the accessing.
The government dimension is where transparency through APIs has the longest and most hopeful history, even as it’s politically contested. Open government data, the Obama directive, budget transparency at the county level via APIs — these represent the use of APIs to make government more transparent and accountable to citizens. APIs are how government data becomes genuinely accessible rather than nominally public, and accessible government data is what lets citizens, journalists, and watchdogs hold government accountable. But this transparency is always politically contested, because transparency constrains power, and those in power have an interest in limiting it. I wrote about how government has benefited from the lack of oversight at social API management layers — a reminder that government is not only a provider of transparency but also a beneficiary of opacity when that opacity serves its surveillance interests. The politics of government transparency through APIs cuts both ways: APIs can open government to scrutiny, and they can also be the infrastructure through which government conducts surveillance with insufficient transparency.
The deeper philosophy I’ve articulated is about why transparency matters at all, and it’s fundamentally democratic. I wrote in 2017 about the reasons we pull back the curtain on technology — because technology increasingly governs our lives, and a society where the systems that govern us are opaque is a society where power operates without accountability. Transparency is how we keep the curtain pulled back, how we maintain the visibility that accountability requires. This connects to the question I raised in 2021 about what “open” really means in the world of APIs — because transparency and openness are constantly co-opted, claimed by platforms that are open enough to capture an ecosystem but not transparent enough to be held accountable. Genuine transparency — real disclosure of how systems work, who accesses them, what they cost, and what they do with data — is a political achievement that has to be fought for against the persistent interest of the powerful in opacity. APIs are one of the most powerful tools we have for achieving that transparency, which is exactly why the fight over how transparent APIs and the systems behind them will be is so consequential. Transparency through APIs is, at its core, about whether the systems that increasingly run our world will be visible enough for the people affected by them to understand, scrutinize, and hold accountable — and that’s one of the most important political questions of the digital age.
References
- Making Transparency Reports Standard Operating Procedure
- API Transparency Report As Essential Building Block
- Keeping A Window Open Into How Power Flows Within Algorithms Using APIs
- If An Algorithm Impacts Our Life It Should Be Opened Up With An API For Auditing
- Transparency In Police Access To Social Platforms Using OAuth And APIs
- The Reasons Why We Pull Back The Curtain On Technology
- Transparency Around Every Company Who Has Access To Our Social Data Via An API
- Government Has Benefitted From Lack Of Oversight At Social API Management Layers As Well
- What Does Open Mean In The World Of APIs
