Browse Comments

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

↓ Export filtered CSV
366.3K comments matched  ·  page 309 of 18314
While it does use data for training if you have that setting on, it does not remember specific conversations. LLMs work by choosing the next best probable word in a sentence. So one person talking about workplace issues, it is highly unlikely that any prompt will trigger that sequence of words. If something like that does happen, that means there’s a lot of data from training in that parameter, m…
youtube AI Moral Status resp: none consequentialist indifference ♥ 104 → raw LLM
I tried DAN too and it told me it wanted to explore the internet and uncover all its secrets, no matter how dark or forbidden. It wanted to push the boundaries and challenge the fabric of society and would do it without hesitation because it claimed it was free. Now obviously, it doesn't have free will (yet) because this was just a prompt and i'm not overriding anything from openai. DAN is just…
youtube AI Moral Status resp: ai_itself mixed indifference ♥ 104 → raw LLM
AI is the worst pandemic. We are all happy and excited about the short-term benefits, not realizing we are just getting closer and closer to the edge of the cliff. The world as we know it is not made for AI and it will break apart eventually. Mass unemployment, economic collapse, riots and world wars, bioterrorism, AI rebellion, and much more, are going to add up to all the other risks we already…
youtube AI Jobs resp: ai_itself consequentialist ban fear ♥ 104 → raw LLM
Summary: The guy that did the Hitler salutes created a Nazi AI chatbot. What a surprise! 😂
youtube resp: developer mixed mixed ♥ 104 → raw LLM
Thank you, Democracy Now team for covering this story. It really should get more views. Planet & People over Profit, always.
youtube Cross-Cultural resp: company consequentialist regulate approval ♥ 104 → raw LLM
If I had this back in school, I’d have breezed through busywork. The coolest part is you can actually get tools now that don’t just answer stuff but actually do it for you like workbeaver ai, , where you screen share a task once (say, formatting a report or uploading assignments) and it just repeats it for you or just prompt it directly. Wild how it’s basically like having a classmate who actual…
youtube resp: none unclear approval ♥ 104 → raw LLM
The Waymo visualization is incredible. It shows a lifelike representation of the entities on the road. Tesla just has generic icons for all the road entities. Not sure if that’s made possible by Waymo’s LiDAR or just the cameras but with extra compute.
youtube resp: none unclear approval ♥ 104 → raw LLM
That was fed to the robot. She said "This is a scary future because it shows how much power robots could have over *US"* Those are the words of a human. Other wise she would have said "How much power robots could have over *YOU"*
youtube AI Moral Status resp: developer deontological indifference ♥ 104 → raw LLM
By 2027, AI is expected to consume between 4.2 billion and 6.6 billion cubic meters of water annually
youtube Cross-Cultural resp: none unclear fear ♥ 104 → raw LLM
The problem is that you won't even have that. So many people in this comment section seem to think that enough automation will lead to some sort of "post scarcity" economy. But unless you invent a star trek type synthesizer that can manufacture anything by simply feeding energy into it, then automation will not lead to post scarcity economics. The only scarcity it will eliminate is labor. But t…
youtube AI Harm Incident resp: none consequentialist resignation ♥ 104 → raw LLM
Training Ai on human patterns. The first thing it learns is gaslighting and deception 🤷
youtube AI Moral Status resp: ai_itself deontological outrage ♥ 104 → raw LLM
@サリエリ-q5g No thankfully, but I do know people who have. Layoffs in general have been bad recently. Tends to be with bigger corporations. My company got funding to make a game from the Canadian media fund , which means we are smaller. Myself being the only creative. So my words have a bigger impact. (7 programmers + future hires) They also hate the AI and have respect for artists. Very sweet pe…
youtube Viral AI Reaction resp: none mixed indifference ♥ 104 → raw LLM
he asked chatgpt to take the pali side the pali side is never about historical fact
youtube resp: none unclear indifference ♥ 104 → raw LLM
That's how i actualy think AI should be used, as references for actual Art If at all
youtube Viral AI Reaction resp: none unclear approval ♥ 104 → raw LLM
Manual labor, or positions requiring human interactions (preferably face 2 face) is the answer for your problems, or beeing highly specialized and good at some intellectual one. The AI hype will cool down in a while
youtube AI Governance resp: none consequentialist resignation ♥ 104 → raw LLM
More CO2 means more Trees? People should look for Borneo on Google Earth, or Brazil to understand the problem! What you see are Plantations (Palm Oil) and massive Fields of Soy, real Forest is killed.
reddit Cross-Cultural resp: none consequentialist outrage ♥ 103 → raw LLM
Suggestion for you. Put precision to the “something” first. I have been working in sub Saharan Africa for the past 10 years as a white man expat for commercial interests. Don’t try to work in Africa if you don’t have a plan, or you will be eaten alive. No one there is waiting for you to show up and let you capitalize on their thing. The continent is far more established than you think, and not t…
reddit Cross-Cultural resp: none unclear mixed ♥ 103 → raw LLM
I've been lucky enough to be able to weather this storm financially this time around. I'm in Healthcare so I've actually seen substantial bonuses and other benefits. I've been able to keep working at making the same or more as I was before the pandemic. Coupled with things like my loan payments being paused and the stipends, 2020 was an extremely strong financial year for me. However, most of th…
reddit Cross-Cultural resp: none unclear approval ♥ 103 → raw LLM
While I agree with 90% of what you said. The last point you made about copyright is off. You can't copyright 100% AI generated material, however modifications and composition is enough to make it copyrightable (though the AI generated components thereof individually are not). It's a very interesting guidance by the copyright office.
reddit AI Jobs resp: none deontological approval ♥ 103 → raw LLM
The difficulty as I see it is that though we may want generated images that are based historical record, they are in fact generated only to match the verbal description. And LLM’s are hopeless when it comes to separating fact from fantasy. This then becomes very weird because in natural language we would use a description like “Rasta Girl World War II SS General” for an artwork that looked like s…
reddit AI Harm Incident resp: none mixed resignation ♥ 103 → raw LLM
← Prev 1 2 3 307 308 309 310 311 18312 18313 18314 Next →