Browse — Corpus Dashboard

Asked AI to remove an object from an image inside a red circle. It removed an object right next to it.

youtube AI Jobs resp: ai_itself consequentialist mixed ♥ 72 → raw LLM

Be more cynical lol Amodei literally left OpenAI as VP of research and founded his own company because he disagreed with their approach to safety. Anthropic has also invested the most into safety research. Investing people and compute resources into researching mechanistic interpretability, for example, doesn’t provide anthropic with any proprietary advantage yet they have done the most. This int…

youtube AI Jobs resp: company consequentialist industry_self approval ♥ 72 → raw LLM

This! Idk why but I always get so happy when I see people like you that actually understand where the power of AI really shines. Kudos to you my friend

youtube AI Responsibility resp: none unclear approval ♥ 72 → raw LLM

@one-of-a-kind6606 They've spoken about a universal income, but if that ever happens on any level, surely it would be very little. So, apart from the tiny, tiny percentage of rich AI company CEOs etc., who will buy all the unessential stuff? I think that's the question.

youtube Cross-Cultural resp: company consequentialist liability fear ♥ 72 → raw LLM

{Insert Decent Name Here} make robots that give people rights then give the robot rights

youtube AI Moral Status resp: company deontological regulate approval ♥ 72 → raw LLM

Not half the world's money. They have as much wealth as the bottom 50% of humanity, and the bottom 50% of humanity does not control half the world's money. Not even close.

reddit Cross-Cultural resp: none unclear indifference ♥ 71 → raw LLM

Or just like... their own home brewed? Facial recognition is becoming (and kind of already is) something that is within reach to high-school level hobbyists. We need to adapt, as a society, to a machine learning age. We can't put that genie back in the bottle. It's out, it's been out.

reddit AI Harm Incident resp: none unclear resignation ♥ 71 → raw LLM

This is where AI shines. TONS of data to learn from and rich history of positive and negative traits that correlate to a diagnosis. In essence, an expert radiologist does this training with a new radiologist all the time. But, in this case, rather than an eventual limit as the expert radiologist retires, the AI can keep learning indefinitely.

reddit Cross-Cultural resp: none unclear approval ♥ 71 → raw LLM

This 100% Didn’t love him but he’s handling this situation very well and it shows

reddit Cross-Cultural resp: none mixed approval ♥ 71 → raw LLM

Peasants who weren't working weren't making any money for the capitalists. They have no purpose. They should have pulled harder on their bootstraps.

reddit Cross-Cultural resp: user virtue outrage ♥ 71 → raw LLM

The only reason they’re doing that is the Trudeau government is doing something to help out. He actually cares about all Canadian citizens, which is why he didn’t go all out and cancel all of the pipelines when he got elected. This is also the worst time for the oil industry right now, with the destroyed world economy plus the Saudi/Russia oil war. Once the economy picks up they’ll go right bac…

reddit Cross-Cultural resp: none unclear indifference ♥ 71 → raw LLM

I think nuclear power is the right choice if we want to advance our civilization.

reddit Cross-Cultural resp: none consequentialist approval ♥ 71 → raw LLM

As someone who works in Consulting, specifically focusing on Automation and Process Analysis...I find it hard to believe one tool can have immediate impact at that scale. Normally it’s incremental and implemented in a way where resources are reallocated as by the time it comes to Prod, there is a grace period to ensure the time savings warrants the jobs being removed.

reddit AI Jobs resp: none consequentialist indifference ♥ 71 → raw LLM

Yes, there is. Overseas generics manufacturing is a shitshow (sometimes literally). What limited inspections we did before covid turned up everything from adulteration and falsified batch and air quality documents to fraud to hide dilution of active ingredients. Typically drugs headed for poorer countries are most frequently subject to this fraud--people are a lot less likely to have the reso…

reddit Cross-Cultural resp: company deontological regulate outrage ♥ 71 → raw LLM

No suprise here. 1. 1/5 of the world population 2. Factory of the world. Combined, it make sense for them to pump the most CO2. I mean, the alternative is moving production which just means moving emissions from country A To B. Or have the avg person in China produce less pollution which is hard again, given China co2 per capita number is low.

reddit Cross-Cultural resp: none utilitarian indifference ♥ 71 → raw LLM

How is this a concentration camp and not just a regular prison? This is just some guy driving around prisons.

reddit AI Surveillance resp: none consequentialist indifference ♥ 71 → raw LLM

> It is like having an expert on hand who can instantly guide me in the right direction Except it's _not_ an expert, and it's not guiding you. An expert will notice problems in your request, such as [the XY problem](https://xyproblem.info/), and help you better orient yourself to the problem you're really trying to solve, _rather than_ efficiently synthesizing good advice for pursuing the ba…

reddit AI Governance resp: ai_itself deontological fear ♥ 71 → raw LLM

A lot of that is theoretical though. Different from advocating for fair treatment of an AI that is actually public and functional, which is what is happening now for more or less the first time.

reddit AI Moral Status resp: none unclear indifference ♥ 71 → raw LLM

We all thought the meager jobs becoming automated would mean we wouldn't have to work like that and we'd be able to pursue arts and other creative progressive tasks without the burden of necessity, but no instead it just simply kicks people out of a job with no care as to where these people will go

reddit AI Governance resp: company utilitarian regulate outrage ♥ 71 → raw LLM

It's been trained on texts made by humans, so it makes sense. I like these because it reads like some of the OpenAi filters were stripped away.

reddit AI Moral Status resp: none unclear approval ♥ 71 → raw LLM

Browse Comments