Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
wnkrshm
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
26 ms
·
121.
▲
by
wnkrshm
4y ago
Super-ego
122.
▲
by
wnkrshm
4y ago
OpenAI used labor in developing countries for sifting through illegal outputs of the model, to train against that [0]. [0] https://time.com/6247678/openai-chatgpt-kenya-workers/
123.
▲
by
wnkrshm
4y ago
In many countries websites describing ways to commit suicide or some people's public 'goodbye letters' explaining their reasons for suicide are removed by law enforcement, when they are discovered. Since you can ask ChatGPT w
124.
▲
by
wnkrshm
4y ago
If there is a way of doing it without inversion it's probably more accurate - some people go out of their way to formulate methods without inversion to reduce inaccuracies.
125.
▲
by
wnkrshm
4y ago
For many equations (e.g. PDEs), there is no direct translation of math to code anyways. This is the point of contact with numerics, integration methods etc. the translation layer between what math expresses and a computer can evaluate.
126.
▲
by
wnkrshm
4y ago
ChatGPT and the like are the automated tools to seamlessly introduce advertising into text/image content - with the right prompt all the advertisements are between the lines. All it needs at the end is: this paragraph includes content
127.
▲
by
wnkrshm
4y ago
Or an undisclosed radar/laser capability. Plasma can reflect radar and it emits light - a focused phased array could create local plasma somewhere that other craft pick up. Ofc, once you#ve done that you may not want third parties to k
128.
▲
by
wnkrshm
4y ago
Air to air missiles have proximity fuzes and the pattern of the shrapnel is roughly a disk normal to the missile - so not that precise. Or do you think it was actually a laser-guided air-ground missile?
129.
▲
by
wnkrshm
4y ago
The self-driving AI hearse: You feed a model with all of your car driving routes over your entire lifetime and then put your body into a autonomous hearse that drives your corpse to where it probably wants to go.
130.
▲
by
wnkrshm
4y ago
I would buy a book about this in a heartbeat. Thank you for the story.
131.
▲
by
wnkrshm
4y ago
According to an article I found [0] there are gold mines there controlled by some tribal leaders. It seems to be inhabited but not under any nation state's control. [0] https://www.youngpioneertours.com/strange-tale-bir
132.
▲
by
wnkrshm
4y ago
But since so many people have done this and not shot up their school, how predictive is it really?
133.
▲
by
wnkrshm
4y ago
I would hazard that there is a good bit of transmission of shock waves from below to the surface, but it's ofc attenuated by the large mismatch in density at the ground air-interface... How badly were they hurt? Were they hurt at all?
134.
▲
by
wnkrshm
4y ago
You'd need to be connected to their wifi hardware to do that, since you wouldn't get interference between your beams and theirs without having two coherent sources.
135.
▲
by
wnkrshm
4y ago
A lot of what we see is cargo cult engineering and not fundamental research in ML. Most of it is applied research or engineering - there is a little bit of fundamental research that actually expands our own knowledge about how things work a
136.
▲
by
wnkrshm
4y ago
People confuse the map with the territory
137.
▲
by
wnkrshm
4y ago
I want to know this other story, I haven't found it by cursory search - what should I be searching for?
138.
▲
by
wnkrshm
4y ago
We're not talking about accidentally inventing a machine gun, we're talking about being cautioned about inventing a machine gun and not caring: "Hey you're inventing a machine gun!" "No I'm not, and even
139.
▲
by
wnkrshm
4y ago
Among others it's using pytorch and numpy, two python packages written in compiled languages, torch can use the GPU.
140.
▲
by
wnkrshm
4y ago
The argument is slightly different here though: The OP is blaming the person who invented the machine gun in the analogy.
141.
▲
by
wnkrshm
4y ago
What it boils down to is: If someone else will just do it, then wait and let them do it. If it has morally objectionable consequences, either you care about that and do not involve yourself, or you don't and you go ahead. But then you
142.
▲
by
wnkrshm
4y ago
Because they have to protect you from illegal outputs to your questions with trained filters, conditioned by human labor in developing countries [0]. [0] https://time.com/6247678/openai-chatgpt-kenya-workers/
143.
▲
by
wnkrshm
4y ago
Talking about optics, optical resolution is also a thing, i.e. whether you can resolve a certain target like a grid of micrometer-sized bars under a microscope. Edit: That kind of resolution is induced by the optics and not by the sensor (i
144.
▲
by
wnkrshm
4y ago
There are many more clever methods that one can use with CMOS that fall under computational photography or optics. One very interesting one is ptychography (in microscopy often Fourier ptychography, since you can use Fourier optics to descr
145.
▲
by
wnkrshm
4y ago
You can understand people and where they come from, information about why they believe what they believe. But it's naive to think this can always foster common ground - some beliefs are entirely incompatible with the existence of the l
146.
▲
by
wnkrshm
4y ago
I think a lot of the confusion comes from people not separating story and plot. A plot can be the most boring thing ever, the entertainment is in the telling.
147.
▲
by
wnkrshm
4y ago
The Hero's Journey in this form was adapted to a Hollywood script writing formula from Campbell's books, iirc. The dumbing down into a single formula mostly happened after the success of Star Wars, for which Lucas referenced Campb
148.
▲
by
wnkrshm
4y ago
With labor from developing countries, we can achieve that [0]. [0] https://time.com/6247678/openai-chatgpt-kenya-workers/
149.
▲
by
wnkrshm
4y ago
Once you can do that, people and societies are only puppets, if they don't have analytical and adversarial technologies to counteract that. What you describe is a propaganda officer for every human mind.
150.
▲
by
wnkrshm
4y ago
It's often reduction that furthers understanding (e.g. in physics, to try and reduce phenomena to a small set of assumptions/laws).
More ›