Reading raw comments is like executing malware on your brain
I found that idea interesting. Will we consider it the norm in the future to have a "firewall" layer between news and ourselves?
I once wrote a short story where the protagonist was receiving news of the death of a friend but it was intercepted by its AI assistant that said "when you will have time, there is an emotional news that does not require urgent action that you will need to digest". I feel it could become the norm.
EDIT: For context, Karpathy is a very famous deep learning researcher who just came back from a 2-weeks break from internet. I think he does not talks about politics there but it applies quite a bit.
EDIT2: I find it interesting that many reactions here are (IMO) missing the point. This is not about shielding one from information that one may be uncomfortable with but with tweets especially designed to elicit reactions, which is kind of becoming a plague on twitter due to their new incentives. It is to make the difference between presenting news in a neutral way and as "incredibly atrocious crime done to CHILDREN and you are a monster for not caring!". The second one does feel a lot like exploit of emotional backdoors in my opinion.
Either it's you deciding as you see it (ie there is no filter), or it's past you who's deciding in which case it's a different person. I've grown mentally and emotionally as I've got older and I certainly don't want me-from-10-years-ago to be in control of what me-right-now is even allowed to see
Just like diet, some people prefer balancing food types and practicing moderation, and others overindulge on what makes them feel good in the moment.
Having food options tightly controlled would restrict personal liberty, but doing nothing and letting people choose will lead to bad outcomes.
The solution is to educate people on what kinds of choices are healthy and what are not, financially subsidize the healthy options so they are within reach to all, and only use law to restrict things that are explicitly harmful.
Mapping that back to news and media, I’d like to see public education promoting the value of a balanced media and news diet. Put more money into non-politically-aligned news organizations. Look closely at news orgs that knowingly peddle falsehoods and either bring libel charges against them or create new laws that address the public harm done by maliciously spreading misinformation.
But I’m no lawyer, so I don’t know how to do that last part without creating some form of tyranny.
isn't that what the upvote/downvote buttons are for? although to be fair, i'd much rather the people of lemmy decide which things are good and interesting than some "algorithm"
There are elements of lemmy who use votes to manipulate which ideas appear popular, with the intention of manipulating discourse rather than having open discussions.
Without wanting to be too aggressive, with only that quote to go on it sounds like that person wants to live in a safe zone where they're never challenged, angered, made afraid, or have to reconsider their world view. That's the very definition of an echo chamber. I don't think you're meant to live life experiencing only "approved" moments, even if you're the one in charge of approving them. Frankly I don't know how that would be possible without an insane amount of external control. You'd have to have someone/something else as a "wall" of sorts controlling your every experience or else how would things get reliably filtered?
I'd much prefer to teach people how to be resilient so they don't have to be afraid of being exposed to the "wrong" ideas. I'd recommend things like learning what emotions mean and how to deal with them, coping/processing bad moments, introspection, how to get help, and how to check new ideas against your own ethics. E.g. if you read something and it makes you angry, what idea/experience is the anger telling you to protect yourself from and how does it match your morality? How do you express that anger in a reasonable and productive way? If it's serious who do you call? And so on.
He's talking about wanting some system to filter out Tweets that "elicit emotion" or "nudge views", comparing them to malware. I looked him up and see he's a computer scientist, which explains the comparison to malware. I assume when he's designing AI he tries to filter what inputs the model gets so as to achieve the desired results. AI acts on algorithms and prompts regardless of value/ethics and bad input = bad output, but I think human minds have much more capability to cope and assess value than modern AI. As such I still don't like the idea of sanitizing intake because I believe in resilience and processing unpleasantness as opposed to stringent filters. What am I missing?
I think you are getting it wrong. I added a small edit for context. It is more about emotional distraction. I kinda feel like him: I want to remain informed, but please let me prepare a bit before telling me about civilians cut in pieces in a conflict between a funny cat video and a machine learning news.
For the same reason we filter out porn or gore images from our feeds, highly emotional news should be filterable
I don't think there's anything wrong with taking a break from social media or news. There are days I don't visit sites like Lemmy or when I do I only click non-news links because I'm not in the mood or already having a bad day. That's different than filtering (as per Karpathy's example) Tweets so that when you do engage it's consistently a very curated, inoffensive, "safe" experience. Again, I only have the one post to go off of, but he specifically talks about wishing to avoid Tweets that "elicit emotions" or "nudge views" and compares those provocative messages to malware. As far as your point regarding blatantly sensationalist news, when I recognize it's that kind of story I just stop reading/watching and that's that.
I WANT to have my emotions elicited because I seek to be educated and don't want to be complacent about things that should make me react. "Don't know, don't care" is how people go unrepresented or abused - e.g. almost no one reads about what Boko Haram is doing in Nigeria (thus it's already "filtered out" by media), and so very little has been done in the 22 years they've been affecting millions of lives. I WANT to have my "views nudged" because I'm regularly checking my worldview to make sure it stays centered around my core ethics, and being challenged has prompted me to change bad stances before. Being exposed to objectionable content before and reassessing is also how I've learned to spot BS attempts to manipulate. It doesn't matter how many times MAGA Tweets tell me that God is upset at drag queens and only Donald Trump can save the world because now I recognize ragebait when I see it. Having dealt with it before, no amount of exposure is going to make me believe their trash and knowing what is being said is useful for exposing and opposing harmful governmental policies/bad candidates (sometimes even helping deprogram others).
The real question then becomes: what would you trust to filter comments and information for you?
In the past, it was newspaper editors, TV news teams, journalists, and so on. Assuming we can't have a return to form on that front, would it be down to some AI?
Why do people, especially here in the fediverse, immediately assume that the only way to do it is to give power of censorship to a third party?
Just have an optional, automatic, user-parameterized, auto-tagger and set parameters yourself for what you want to see.
Have a list of things that should receive trigger warnings. Group things by anger-inducing factors.
I'd love to have a way to filter things out by actionnable items: things I can get angry about but that I have little ways of changing, no need to give me more than a monthly update on.
Most recent Ezra Klein podcast was talking about the future of AI assistants helping us digest and curate the amount of information that comes at us each day. I thought that was a cool idea.
It makes a lot of sense. It also presents an opportunity to hand off such filtering to a more responsible entity/agency than media companies of the past. In the end, I sincerely hope we have a huge number of options rather than the same established players (FANG) as everything else right now.
Our mind is built on that "malware". I think it's more accurate to compare brain + knowledge to our immune system: the more samples you have, the better you are armed against mal-information.
This sounds like the theories that were more prevalent before germ theory. Surgeons or obstetricians would argue that washing hands was a disservice to the organisms they get into.
Immune systems still get sick and can be overwhelmed. There is a mental hygiene that needs to exist.
But that leaves out the psychological effects of long-term exposure to ideas. If you know for a fact that the earth is round, and for the next 50 years all the media you consume keeps telling you that the earth is flat, you will at some point start believing that (or at least become unsure).
Every piece of information you receive has some tiny effect on you.
We are already having tons of filters in place trying to serve us information we are interested in, knowledgeable enough to digest, not-spammy, in the correct language, not porn or gore, etc... He is just proposing another interesting dimension. For instance, I am following AI news and news about the Ukraine conflict but I prefer to keep them separate and to not be distracted by the other when I get my fill on one.
The only way I found with Twitter (and now Mastodon) to do it is to devote twitter only to tech news.
I don't think he is proposing another dimension, but rather another scale. As you already said, we already filter the information that reaches us.
He seems to take this idea of filtering/censorship to an extreme. Where I see filtering mostly as a matter of convenience, he portrays information as a threat that people need to be protected from. He implies that being presented with information that challenges your world view is something bad, and I disagree with that.
I am not saying that filtering is bad. I too have blocked some communities here on Lemmy. I am saying that it is important not to put yourself in a bubble, where every opinion you see is one you agree with, and every news article confirms your beliefs.
Kind of, but the guy being a prominent LLM researcher, it kind of hints at the ability of not inflicting it on humans nor suffering from having to design an apolitical structure for it.
I think most people already have this firewall installed, and it's working too well - they're absorbing minimal information that contradicts their self-image or world view. :) Scammers just know how to bypass the firewall. :)
We already have a firewall layer between outside information and ourselves, it’s called the ego, superego, our morals, ethics and comprehension of our membership in groups, our existing views and values. The sum of our experiences up till now!
Lay off the Stephenson and Gibson. Try some Tolstoy or Steinbeck.
Reading, watching, and listening to anything is like this. You accept communications into your brain and sort it out there. It's why people censor things, to shield others and/or to prevent the spread of certain ideas/concepts/information.
Misinformation, lies, scams, etc function entirely on exploiting it
Not really. An executable controlled by an attacker could likely "own" you. A toot tweet or comment can not, it's just an idea or thought that you can accept or reject.
We already distance ourselves from sources of always bad ideas. For example, we're all here instead of on truth social.
I think you're too optimistic as to how difficult it is to influence people. Just think of the various, obviously false, conspiracy theories that some people still believe. I think that for every person there is some piece of information/news, that is just believable enough without questioning it, that is going to nudge their opinion just ever so slightly. And with enough nudges, opinions can change.
Leaving aside the dystopian echo chamber that this could result in, you could argue that this would help with fake news by a lot. Fake news are so easy to spread and more present than ever. And for every person there is probably that one piece of news that is just believable enough to not question it. And then the next just believable piece of news. and another. I believe no one is immune to being influenced by fake stories, maybe even radicalized if they are targeted just right. A firewall just filtering out everything non-factual would already prevent so much societal damage I think.
There are enormous issues with who decides what makes it through the filter, how to handle things that are of unknown truth (say ongoing research), and the hazards of training consumers of information to assume everything that makes it to them is completely factual (the whole point of said fake news filter). If you'd argue that people on the far side of the filter can still be skeptical, then just train that and avoid censorship via filter.
Yeah, I agree. it's not easy to determine truth, and whoever decides truth might introduce bias that then gets rolled out to everyone. With ongoing reserach or unknown information, you could just have a "currently being researched" or "not confirmed yet" attached to the information. I'm just saying that in an ideal world where this does work, it could be safer than relying on people being skeptical, because everyone fails to be skeptical about something eventually.
We already have a firewall its our thoughts. The information can nudge us but it's fighting an uphill battle against everything we already know and believe.
Your thoughts (I guess you mean your past knowledge, intelligence and critical thinking) allows you to dismiss lies, but it does not shield you from the emotional charge of some news.
I remember watching a video from a psychiatrist with eastern Monk training. He was explaining about why yogis spend decades meditating in remote caves - he said it was to control information/stimuli exposure.
Ideas are like seeds, once they take root they grow. You can weed out unwanted ones, but it takes time and mental energy. It pulls at your attention and keeps you from functioning at your best
The concept really spoke to me. It's easier to consciously control your environment than it is to consciously control your thoughts and emotions.
It's definitely an angle worth considering when we talk about how the weakest link in any security system is its human users. We're not just "not immune" to propaganda, we're ideological petri dishes filled with second-hand agar agar.
Perhaps we can establish some governmental office for truth that decides whether any shitpost can be posted without the sterilization and lobotomization of the poster
Or maybe some kind of "community value" score for people with the right thinking
I've thought about this since seeing Ghost in the Shell as a kid. If direct neural interfaces become common place, the threat of hacking opens up from simply stealing financial information or material for blackmail; they may be able to control your entire body!
You are responsible for what you do with the information you process. You're not supposed to just believe everything you read, or let it affect you. We don't need some government or organization deciding what can be shown online. Read history and see what follows mass censorship.
I am bewildered that so many people contrive this as suggesting it should be a government or a company deciding what to show you. Obviously any kind of firewall/filter ought to be optional and user controlled!
In fairness social media already has a problem with creating an echo chamber that serves only to reinforce and exaggerate your existing world view. I think to some extent exposure to perspectives other than one's own is important and healthy.
Yes, lemmy too is that. We need to meet people and then form groups online. I had devised a solution for exchanging public keys in person and verifying each content thereafter with that key.
I wonder if maybe it's more apt a comparison to say that allowing raw comments to affect you in a strong way is like running a random program as root. To a certain extent you have to let this kind of harmful content in.
P.s. the short story sounds cool - is it available to read anywhere?
I look forward to factchecker services that interface right into the browser or OS, and immediately recognize and flag comments that might be false or misleading. Some may provide links to deep dives where it's complicated and you might want to know more.
I mean, this is just called censorship. We censor things for kids and all kind of people in or lives all the time. We censor things for ourselves when we don’t feel like reading the news or opening a text from a specific person. This is not some novel concept.
Not really. This is user-controlled filtering. Censorship is done to push a specific worldview to victims. Filtering we do it all the time for spam for instance.
But the post is explicitly about Tweets that challenge emotions and views and how that's harmful. It's one thing to want to see fewer suspicious offers from Nigerian princes and horny MILFS in my area. It's another to tell an AI that you don't want to see events or conversations that might be upsetting or make you think about ethics, politics, etc.
P.S. I'm replying to you a lot today, just want to say I'm not trying to be abusive or follow you around. You keep making points on this page that I want to engage with, and hopefully it's not coming across as persecution.
I've, since I was young, had mantra. If you don't why your are doing something someone else does. Its not always conspiracy or malicious its literally the basis of the idea of memetics, shareable and spreadable ideas that form the basis of who we are and what we know.