Welcome to another issue of CTPI Insights, your premium newsletter on technology design and policy.

Today, we'll cover:

The Systemic Approach to Technology;
Simplistic perspectives that conceal real clashes of fundamental rights in tech debates;
Fresh new CTPI publications just off the (digital) presses and much more.

Let us just tell you: we've been busy. Today, we are announcing the publication of three brand-new CTPI Frontiers articles. Two are the unabridged version of our contributions to the Forum Pix about payment managers and “Pix Automático”. They are both in Portuguese and are intended for our subscribers only. But let us know if you need them in English and are having a hard time translating them as they are. We might be able to help.

The third publication is something we've been working on since we first started CTPI. It's an article about how we believe we should think about, interact with, and build technology. It has a little philosophy, a little computer science, and a lot of our hopes and dreams for the future. You can think of it as CTPI's very own manifesto. Feel free to share that one with whomever you like, don't forget to share your thoughts with us at any time, and stay safe, everyone.

Now, let's get to it.

M. and J.

Introducing the Systemic Approach to Technology

Today, we proudly announce the publication of “Systemic Approach to Technology (SAT): The Case for Society-Focused Technology”. It results from a long debate and reflection about the thought process involved in developing lasting technical solutions.

You can read the full paper here. But to give you a quick taste of it for your weekend, we put together a short and easy summary of what we now call the Systemic Approach to Technology. If you're intrigued, read on and save the full SAT article for your next reading session.

Why?

The Systemic Approach to Technology is useful throughout the entire product lifecycle. It helps engineers, product managers, designers, and executives make better decisions. Regulators and policymakers can also use it to understand market forces and influence behavior for the common good. When citizens and society learn how technology works, we can have more intentional discussions about what we want and influence its development.

How?

Step 1: identify premises

First, we should practice viewing all observed events as manifestations of an underlying system. Events can be the state of low competition in a market, the speed of adoption of a given product, etc.

We do this by working backward from the event we can observe to their underlying premises (that is, all the elements and interactions between elements that could result in the event we observe). This investigation into the premises of an event can be broad when exploring multiple causes or deep when examining the logical assumptions behind a specific premise (the premises of a premise).

Step 2:

As we identify the event's potential premises, we must make an additional analysis. For every new premise we add to the list, we must ensure it is compatible with all the previous ones. If they are not, we must either adjust the new premise or the ones we believed were true before. By doing this, we develop a mental model of the system that could potentially produce the observed event. It's the compatibility of underlying premises that makes sure our model of the system can explain all its events.

Step 3

Next, whenever a new event occurs, we must compare it with the mental model we have created. We need to determine if the new event can be explained by the model of the system we have envisioned. If it can, that's great. If not, we need to modify our mental model so that it can be compatible with all observed events.

So what?

So, we have a mental model of a market or any other system. Now, what can we do with that? Nothing short of unlocking three superpowers.

Superpower 1: future telling

Once we have created a model for the system and its inner workings, we can make a scenario analysis by changing one or more parts of the system. Since we know how the elements of a system interact, we can reasonably predict how these changes might impact other elements and their interactions. This type of thinking helps us make predictions about the effects of new legislation, the introduction of new products to the market, and the potential impact of new technology applications.

Superpower 2: x-ray vision

The second way to use systemic reasoning is to actively seek opportunities to make changes within a given system. Product teams and executives can also utilize systems thinking to make strategic decisions after analyzing the market system in which they operate.

Superpower 3: creating your own reality

Finally, systems reasoning can be used to design a new system from scratch as a means to make an impact in a broader system. For that to happen, technologists need to take specific demands for use cases as manifestations of an underlying system and take a higher degree of abstraction to identify its most basic elements and learn how they interact with each other. And then, create a model that could represent the elements and their interactions to figure out what levers could be used to bring equilibrium to the system. Finally, it is the technologists’ job to model the underlying system in a way that primitives and rules can be used to describe not just the solution to the demand at hand but an unknown number of other demands that can result from the same system.

Who defines acceptable risks?

The short answer is: that society should. In a previous update, we argued:

We as a society must regain the focus and the locus of the discussion. We must learn the language of this debate and claim the position of protagonists in this process so we can make sure our long-term interests are truly at the heart of every decision moving forward.

Unfortunately, there is a belief that simplistic technology solutions can address society's complex issues. From Joana Moll:

The belief in a simple technological solution to society’s complex problems has taken root. This techno-solutionist trend has proved ineffective in offering solutions, but it continues to shape the way we see the world. The design of digital technologies accordingly suppresses our ability to imagine other realities.

And the discussion on CSAM proliferation is full of simplistic pseudo-solutions. Consider this photo:

Campaign in NYC urging Apple to improve CSAM detection. Photo credits: Jonas de Abreu.

The image above is from a campaign by the HEAT initiative. It's a collective effort of concerned child safety experts and advocates encouraging leading technology companies to actively detect and eradicate CSAM, particularly images and videos, on their platforms.

In an open letter addressed to Tim Cook, they wrote (the emphasis is ours):

Child sexual abuse images and videos must be treated differently than other content and requires a different solution because such images and videos are illegal and evidence of the rape or molestation of children. 60% of all reported child sexual abuse material features children under the age of 12.2 In our recent research, we have come across hundreds of cases of child sexual abuse that have been documented and spread specifically on Apple devices and stored in iCloud. Had Apple been detecting these images and videos, many of these children would have been removed from their abusive situations far sooner.

This campaign and open letter are a reaction to Apple's decision not to deploy NeuralHash, a client-side content scanner that would bypass their iCloud end-to-end encryption to detect and decrypt CSAM. That way, authorities could gain remote access to the personal files of millions of iCloud users to investigate this kind of crime.

This is not the first time we talk about this. We mentioned a few times before the crusade of British policymakers against encryption in the name of law enforcement. Last month, the U.K. Online Safety Bill was approved. From Forbes:

This new legislation seeks to protect people from illegal content and activity online, with a critical focus on child protection so that they do not encounter content and activity that could be harmful to them. (…) Broadly speaking, the platforms will have to perform risk assessments to understand the extent to which their users will encounter illegal or harmful content, block access to this content and arbitrate between other interests that might be engaged by blocking, such as journalistic interests and freedom of expression.

Here, as well as in the case of the campaign mentioned above, we see the same problematic pattern emerge. The way the discussion is framed ignores the complexity of the issue. It looks like big tech companies just need to build new technology, and the children's online safety problem is solved. A simplistic, techy approach that ignores how complex CSAM fighting and prevention is will cause damage and, more likely than not, fail to solve the problem.

Of course, detecting CSAM and removing it from the internet is a noble cause. It is illegal and horrible, and the internet would be better and safer for everyone without it. There's no question Apple and the other platforms fully agree as well. But to move closer to our common goal of making the internet safer for all, we need a path that embraces complexity instead of ignoring it.

Embracing complexity

At the deepest level, the conflict between End-To-End Encryption (E2EE) and CSAM fighting is the conflict of two fundamental rights: the right to privacy and the right to physical and emotional security. It is pretty clear that CSAM violates a right to physical and emotional security and the special protection childhood enjoys. Although it may not be as apparent at first glance, the fact is that encryption is, in today's world, indispensable to privacy and other correlated rights, like individuality, or even life, liberty, and personal security. From UNESCO's Human Rights and Encryption publication:

There needs to be recognition of cryptographic methods as an essential element of the media and communications landscape. What ultimately matters, from a human rights perspective, is that cryptographic methods empower individuals in their enjoyment of privacy and freedom of expression, as they allow for the protection of human-facing properties of information, communication and computing. These properties include the confidentiality, privacy, authenticity, availability, integrity and anonymity of information and communication.

While the publication does not directly cover E2EE, that is the encryption technique that can provide the highest levels of privacy to electronic communications. And with a big part of our lives relying on electronic communications, E2EE is fundamental for privacy. Without E2EE, our communications can be read and misused by hackers, governments in mass surveillance programs, or even the very employees from these platforms.

Arguing against E2EE encryption (or encryption altogether) to protect children (or other law enforcement goals) frames the discussion as if it was a conflict between children's protection and the financial interests of big corporations. However, that simplistic perspective is both wrong and dangerous. It disguises what is really at stake in this kind of discussion, which is the conflict between two sets of fundamental rights.

Since E2EE is fundamental to privacy, banning E2EE would not be a victory for human rights over greedy executives. It would mean removing one of the most essential tools to protect everyone's rights, including children's.

To help in this and many other discussions, we have published our approach on how to build technological solutions that serve society instead of trying to force society to conform to a tech solution. The Systemic Approach to Technology is intentionally designed to work on complex problems instead of ignoring the complexity.