Thank you for an interesting question.
I apologise in advance for this lengthy response and for using English. In the interest of minimising any disruption caused I encourage anyone wishing to reply to stick with Czech even if you are comfortable expressing yourself in another language. I'll (just about) manage!
Nějaká část problému by se měla řešit v současnosti probíraném Zákonu o digitálních službách (DSA)
What aspect(s) specifically?
Napadá vás, co dalšího s tím dělat?
A root cause analysis?
The issue may be dissected in the following parts:
- 1. Facebook (allegedly, etc.) knowingly promotes divisive or controversial content in order to keep users on its sites.
- 2. At least some of that content is hate speech or otherwise violates fundamental rights or people's dignity, or patently threatens their safety. Let us call this "malign content".
- 3. Controversial content per se is not an issue, on the contrary. Let us refer to this as "controversial content".
- 4. The first question would be where to draw the line between one and the other. That in itself is already a colossal and delicate task. Much has been written about it and there are different schools of thought on the subject.
Please note: since we are looking at it from a Western perspective, John Stuart Mill's 1859 treatise "On Liberty" (out of copyright and available from any of a number of sources) would seem like a reasonable starting point.
Furthermore, if we acknowledge the heavy influence of U.S. cultural and social values in our own contemporary European society, I figure an awareness of Scanlon's work on the subject would not go amiss (e.g., https://doi.org/10.1017/CBO9780511615153), keeping in mind the evolution of his views over the years, and the reasons why.
In fact, the referenced work includes an essay on the very subject of content regulation – from 1990! In a nutshell, his conclusion is: it's difficult, and it's illusory to believe that it can be done in a way that does not involve "large and obvious elements of value judgment".
- 5. The first question would then be how to tackle "malign content" without negatively impacting on the merely controversial. We need to do this because otherwise we would be limiting freedom of expression beyond what is acceptable or judicious.
- 6. As per Scanlon's conclusion referencing the need for value judgment, it seems apparent that this could not be done solely or primarily by algorithmic application, whether the algorithm is executed by a machine or a human, such as someone employed as a content moderator who is limited to following a script.
Algorithms may help in flagging content (or more generally, acts of expression) for attention, but they cannot be used for censorship. This is, I believe, clearly demonstrated by the many instances of misguided application, such as where sexual education campaigns or classic works of art were "banned" by an algorithm, and it is for this very reason (amongst others) that the Digital Services Act is fundamentally flawed in its requirement for "expeditious" removal of information as it does not explicitly exclude the possibility of automated removal.
- 7. Facebook is but one actor, but the issue of tolerating, promoting or profiting from "malign content" is one that concerns every so-called social media provider out there, whether corporate or grassroots, garden-walled or federated, and it does not affect only, or even primarily, Western societies (https://doi.org/10.1145/3292522.3326034, preprint available from the arXiv).
- 8. Disinformation, misinformation, propaganda and deception (aka "fake news") are documented since at least ca. 500 BCE (e.g., Sun Tzu) and given that analogous behaviour is present in other organisms in nature it is reasonable to conclude that it serves an evolutionary purpose, at least in some forms. This means that any attempts at getting rid of it are futile.
- 9. From that perspective, the ability to detect, analyse, understand and contextualise falsehoods would appear to be of primary importance.
- 10. At the same time, reacting to "malign content" by limiting its spread and/or its effects also has its place but, in my view, it is harmful and ineffective to enact this as the first line of defence. This only works at a tactical level.
- 11. It would be useful to understand what has changed (or not) with the advent of the internet on the one hand and large content providers on the other hand.
A priori the dynamic would be similar to, or merely a continuation of, the printing press revolution: its effect was to make information and knowledge available to a wider audience (sound familiar?), faster and cheaper than ever before; it made generalised literacy possible and, through that, became a main enabler of the industrial revolution, bringing us to where we are today.
But, as technology does not know about morality, it was also used to disseminate disinformation and propaganda, as well as information and counter-propaganda (see the Elders of Zion hoax as an example of the former and Émile Zola's response to the Dreyfus affair for the latter).
And this is where we seem to be today, except bigger and faster.
So, what do we actually do about it?
It would be very arrogant of me to imply that I have any solutions, which of course I do not. But I would like to put forward the following ideas for consideration:
a. Move cautiously, avoid knee-jerk reactions and all overreaction.
b. Appreciate the full scale of the issue. This is not (just) about Facebook, social media, or even the internet.
c. Recognise that the same technology that is used for the dissemination of "malign content" is also used in many more positive ways. We wouldn't want to be the music industry in the 90s and go lobbying for a CD-ROM tax because someone might use it to infringe copyright.
d. Admit that "malign content" is not and never will be fully eradicable and incorporate that principle in any defensive strategy designed to mitigate its impact.
e. Limit any government-imposed censorship to the minimum necessary to protect fundamental rights and the safety and dignity of people and communities. Do encourage swift action in those cases.
e'. Ensure that government itself is excluded from the above protections. Obviously, this exclusion does not extend to the individuals that constitute a government, only to the institution itself.
f. In my view and experience, educating people in critical thinking from the youngest possible age, ideally as soon as they are able to read with some fluency, is perhaps the single most important way of preventing the worst effects of exposure to "malign content". Of course, this will do nothing for the Facebook generation, but it might help the TikTok generation and those who will come after that.
g. The above can only fully work in the context of a tolerant, diverse and well educated society. I appreciate that this is a bit of a chicken and egg situation.
Máte nápad, na co se jí případně máme zeptat?
Would she have any suggestions on how could discussion be conducted in a way that safeguards the principles mentioned above (respect of fundamental rights, dignity, safety of individuals and communities) while protecting, and even encouraging, debate on controversial topics? Is there anything that, in her opinion, governments should do or refrain from doing?
On a side note: if Facebook have indeed been manipulating content placement and visibility as alleged, aside from committing copyright violation (under 17 U.S.C. § 512) are they not also engaging in editorial work and open to liability for the content so selected?
Once again, thank you for an interesting subject and apologies for the verbiage.
