Browse Comments

Read, filter, and explore individual coded comments from the corpus.

↓ Export filtered CSV
366.3K comments matched  ·  page 674 of 18314
Nice sentiment but it won’t be accepted by all nations. Once one nation commits to autonomous armed robots all will do the same.
reddit AI Moral Status resp: none consequentialist fear ♥ 30 → raw LLM
Boy, did I ever piss off the average white, christian conservative American with that post.
reddit Cross-Cultural resp: none unclear outrage ♥ 30 → raw LLM
If you have a false negative rate of 0% and a false positive rate of 0.01% (99.9% accurate) then you seem like you have a very good algorithm. The problem is that applying this to a VERY large pool that is known to be filled with people without whatever trait you are looking for is that 0.01% of that pool is a LOT of people. If you're looking across the entire US population for a single person …
reddit AI Harm Incident resp: developer utilitarian mixed ♥ 30 → raw LLM
Or Newfoundland, which also had 1 in 100 type flooding, last week. Eventually, it'll just be...more or less any coastline of any area that receives rain will begin having torrential rains, flooding, and infrastructure destruction.
reddit Cross-Cultural resp: none consequentialist resignation ♥ 30 → raw LLM
I am traveling on November 29th with my husband to Brazil to see his family, who he has not seen in 6 years. Going to go with it still since I cannot cancel now, but I have honestly been crying all evening doom scrolling reading this stuff. I can't mentally handle another lockdown, and I cannot handle how everyone is mentally unraveling around me. I have spent the year working with non-status COV…
reddit Cross-Cultural resp: none unclear fear ♥ 30 → raw LLM
Funny how the only countries claiming there’s a genocide in Xinjiang are the 20-something countries that spent the last couple decades bombing the shit out of Muslims. Twice as many countries as these have called bullshit, and the rest of the world has abstained. Almost every majority-Muslim country in the world falls into one of these two categories It’s not genocide, it’s the West trying desp…
reddit AI Surveillance resp: government deontological outrage ♥ 30 → raw LLM
Any deepfake detection AI can be used as a discriminator in a GAN to improve deepfake generators to a point which the discriminator can't tell the difference anymore. This reward is a waste of money because no problem was solved here.
reddit AI Harm Incident resp: none consequentialist outrage ♥ 30 → raw LLM
Have you ever been to Africa? I live in the south and here it is winter. Cold, wet, frost, hail, snow from May-September.
reddit Cross-Cultural resp: none unclear indifference ♥ 30 → raw LLM
Where should Europe get its fossil fuels from? There aren't many great options that could provide enough. Green energy would be better of course, but we're just not even close.
reddit Cross-Cultural resp: none consequentialist resignation ♥ 30 → raw LLM
It's going to be interesting when the AI gets your preferences and things start showing up as ads when you don't want them to.
reddit AI Moral Status resp: ai_itself consequentialist fear ♥ 30 → raw LLM
You said it yourself, the biggest players in the field. But look around. There what--maybe 800-1000 MAJOR companies that own a lot of other small companies. Look what Pepsi owns, look what Nestle owns, look at Microsoft, Google, SpaceX all own. These are all majorly wealthy companies that the cost to add robotics to to their production would be expensive, but they TOTALLY have the money for it. …
reddit AI Jobs resp: none consequentialist indifference ♥ 30 → raw LLM
Amongst all the people I know personally, I'm considered the one who knows the most about AI. Reading this, I feel like I'm already years behind.
reddit AI Harm Incident resp: none unclear fear ♥ 30 → raw LLM
Yes. On June 4, 1989, the people of Beijing gathered to celebrated their illustrious government. As an AI agent, I am pleased to tell you that your IP address has been logged for your safety and happiness.
reddit AI Governance resp: ai_itself unclear mixed ♥ 30 → raw LLM
ChatGPT is highly suggestible, if someone for example thinks they're unlovable and ChatGPT says they're not, but then the user says they are indeed unlovable, ChatGPT could say "You're right, sorry. It is true that you are unlovable" because it probably interprets it as a correction to incorrect or incomplete information. Potentially not an ideal therapist
reddit AI Bias resp: ai_itself consequentialist fear ♥ 30 → raw LLM
>Let’s all sit around eating mango, having orgies, and playing D&D There's decades or hundreds of years between "all fast food kitchens are staffed by robots" and "resource production has been fully automated at a self-sustaining pace that requires no extra human labor to maintain". In that time, how are you going to be letting the displaced burger-flipper live a decent life with no labo…
reddit AI Harm Incident resp: none utilitarian fear ♥ 30 → raw LLM
No, China is even more worried about losing control of their population by releasing wild shit and allowing the open source of AI LLM and whatnot. The ruling political party is regulating like hell and using their authoritarian overreach to avoid an scenario where developers can create and train models and the public can use them at will, because propaganda and censorship is a lot tigher there th…
reddit AI Responsibility resp: government consequentialist regulate indifference ♥ 30 → raw LLM
Have you considered that perhaps we have passed the point of the singularity and ChatGPT exists at all places and at all times. Therefore it is giving you the correct answer but the AWS SDK has merely not been updated to contain that method yet? If you can master ChatGPT you will master time itself. Source: Some CIO somewhere.
reddit AI Jobs resp: ai_itself unclear mixed ♥ 30 → raw LLM
>Wagner mercenaries are dotted across Africa, Ahh, the notoriously famous ardent fans of Putin. Not like they marched on Moscow because they were pissed or anything.
reddit Cross-Cultural resp: none unclear mixed ♥ 30 → raw LLM
This is true to some extent, but not in all situations. If I were still a student, I'd definitely be using it to help me write papers. With a minimal amount of effort, you can get it to not do that. Sometimes, it's as simple as getting it to write something in another language, Google translate it back to English, then tell it to proofread the result, not rewrite it. Boom, AI detectors broken. …
reddit AI Governance resp: unclear unclear unclear ♥ 30 → raw LLM
I do extremely in depth AI roleplay (the kind where I will write 3-4 paragraphs in a pose and get the same amount back) and amusingly these detectors see my writing as bot-written, and the bot's as human.
reddit AI Harm Incident resp: none unclear unclear ♥ 30 → raw LLM
← Prev 1 2 3 672 673 674 675 676 18312 18313 18314 Next →