#ResearchEthics #DigitalEthics “Community-level experiments impact communities, not just individuals.”

For me, one of the most important sentences in an overall thoughtful declaration by reddit moderators (available here https://www.reddit.com/r/changemyview/comments/1k8b2hj/meta_unauthorized_experiment_on_cmv_involving/)whose community was not only studied by UZH researchers, but also unethically manipulated by their undisclosed, egregious AI-bots.

More infos [in German] here: https://dnip.ch/2025/04/28/reddit-user-mit-ki-bots-getaeuscht-das-fragwuerdige-experiment-der-universitaet-zuerich/

Ok, a few of my thoughts on the infamous UZH Reddit-CMV AI experiment are now on the record🤖

Thanks to https://chaos.social/@adfichter for publishing my perspective on DNIP (in German): https://dnip.ch/2025/05/01/unautorisiertes-reddit-ki-cmv-experiment-verdaechtiges-institut-uzh/

[ETA correct user name]

adfichter 🖋 (@adfichter@chaos.social)

1.97K Posts, 524 Following, 3.61K Followers · Swiss Investigative Journalist of the Year 2020, 2021 and 2024. Investigative Tech-Reporter Republik & Dnip.ch Interests: #Netzpolitik #Privacy #InfoSec #DigitaleEthik #Technology And: #cats #chess #music #redwine

chaos.social

Obviously, the deception and disregard for community norms is shocking. In addition, from a big picture view, it also hints towards another huge problem: academic research is concerned with ethical standards, whereas commercial platforms can analyze and influence their users without any ethical oversight, BUT: this does by no means excuse unethical research. On the contrary, it underlines how urgently we need more ethics and care in all digital spaces!

(in German: https://dnip.ch/2025/05/01/unautorisiertes-reddit-ki-cmv-experiment-verdaechtiges-institut-uzh/ )

Unautorisiertes Reddit-KI-Experiment: Ein verdächtiges Institut an der Universität Zürich - Das Netz ist politisch

Seit unserer Publikation am Montag zum geheimen Reddit-KI-Experiment rund um die Universität Zürich hat sich einiges getan.

Das Netz ist politisch

Interestingly, I have learned in another context, that Swiss journalists have a hard time finding researchers at Swiss institutions willing to comment on the record -- which surprises me and I wonder why. (Because of ill-placed collegiality? Because of UZH's institutional power? Or ...?)

Isn't it important to have conversations about both research ethics and digital ethics, including publicly? I fear that avoiding these might errode even further the already fragile public trust in academia...

@annajobin I agree.

OTOH, I understand that people fear being considered a “nestbeschmutzer,” especially in a small scientific community in a small country. There’s a small number of key players that have a lot of leverage through learned societies, the academies, the SNSF, the SERI.

And if it concerns your own institution, it could be interpreted as a violation of your duty of loyalty…

The question is whether there’s a real risk of retaliation, or whether the fear is unfounded.

@mxp Thank you for your thoughts, and I understand that it’s different when you know the involved people personally. If I understand you correctly you hint at a general sort of “solidarity” that some academics may demand from each other? I am all for solidarity, but rather against unjust and unjustified attacks, and this is not that. Right? (Also: Fear of retaliation should always be unfounded because retaliation should never happen!)

@annajobin I guess you could call it “disciplinary solidarity.” I totally agree, disciplines should be interested in upholding standards and maintaining integrity (that's their purpose!), thus misconduct should be addressed rather than covered up. But like in other types of organization, criticism of individuals—even when justified and constructive—might be perceived as an attack on the whole organization. “Don't rock the boat!”

Of course, retaliation SHOULD NOT happen, but it certainly does.

@mxp Thanks, I see we agree.