ganzuul

20 ms·

241.
by ganzuul 3y ago
It has to be very hard to overcome the bad vibes of being in a situation like this. The technology seems indecipherable to a non-techie. The law seems indecipherable to a layman. The ethics seem indecipherable to everyone. With so much conf
242.
by ganzuul 3y ago
Just an opinion piece, no information.
243.
by ganzuul 3y ago
House plant for me. Carnivorous.
244.
by ganzuul 3y ago
Reminds me of speech inflection.
245.
by ganzuul 3y ago
I feel like we brought this on ourselves when we started autocompleting replies to questions like "How are you?" with, "Fine, thank you." instead of thinking about it and giving honest answers.
246.
by ganzuul 3y ago
Bhakti is much faster.
247.
The First Precision Lathe – Constructing the Antikythera Mechanism(youtube.com)
19 points by ganzuul 3y ago|1 comments
248.
by ganzuul 3y ago
It would be interesting to have a taxonomy of planning problem hardness. A lot of the references are about control problems, which AFAIK are rather deterministic. Making plans when dealing with incomplete or missing information requires str
249.
by ganzuul 3y ago
This is really quite lovely. It should encourage a lot of self-expression. Any plans for tablets? Kids seem to use those more than PCs these days.
250.
by ganzuul 3y ago
Is it performant? From the description it seems like it should be AGI if it is.
251.
by ganzuul 3y ago
That's what Spore promised but did not deliver.
252.
by ganzuul 3y ago
Makes me feel like the current timeline isn't right.
253.
by ganzuul 3y ago
In identifying categories and repopulating them Gemini Pro seems to be slightly more diverse than GPT-3.5-Turbo, while the latter tries harder to stay within the constraints given.
254.
by ganzuul 3y ago
Information warfare, measuring and predicting everything leads to a lifestyle without peace or satisfaction. It's a corporate lifestyle which we should defend against. It does offer some quick thrills but they are not sustainable. It i
255.
by ganzuul 3y ago
Datasets generated out of GPT-4 used to train open source models have been referred to as synthetic. Before LLMs there was a trend of playing classic video games with deep learning, so perhaps the next datasets will come out of more modern
256.
by ganzuul 3y ago
Guessing they will be used to generate synthetic data from simulation to feed the next gen of AI. No copyright issues then. There might be room for clubs of AI hobbyists who curate datasets and build simulations.
257.
by ganzuul 3y ago
Ad personalization is also a financial model custom made to exploit you.
258.
by ganzuul 3y ago
It would be interesting to know if suicidal signals increase when young people are exposed to advertising.
259.
by ganzuul 3y ago
I don't think poking them will lead to much understanding of intelligence.
260.
by ganzuul 3y ago
I mean to say that since so few bits are required to encode perception, the universe in which Boltzmann Brains exist in is teeming with intelligence. Poetically, an intelligent universe seems closer to reality thanks to this new observation
261.
by ganzuul 3y ago
In classical terms, out of pathos, logos and ethos, LLMs are logos. They discover metrics. https://en.wikipedia.org/wiki/Metric_space Neural networks are an optimization which exploits structure in sparsity. Intelligen
262.
by ganzuul 3y ago
We do need to clarify what our language means. Is this self-awareness or the hard problem of consciousness?
263.
by ganzuul 3y ago
This is an interesting point of fact about what Boltzmann Brains could be.
264.
by ganzuul 3y ago
> Ultimately, the burden of proof is on those who propose that AI models are conscious to present an argument in support. Not a fair proposal. Not how consciousness works. You can only prove your consciousness to yourself.
265.
by ganzuul 3y ago
Purse my dreams like my life depends on it.
266.
by ganzuul 3y ago
How does a child accounting for a wild dream measure up in this test?
267.
by ganzuul 3y ago
It would be nice to see the research paper to figure out how advanced their model is and what previous research they are referencing.
268.
by ganzuul 3y ago
Please excuse my presumption. It must have been challenging! I think the next generation will leave us in the dust if they get to tap into the potential that is there.
269.
by ganzuul 3y ago
You are also privileged not to suffer from dyslexia, ADD, ADHD, and so on. Add all the letters up and we see that so many children are affected that we should probably be looking elsewhere for their talents. The idea that we could actually
270.
by ganzuul 3y ago
Maybe AP an Reuters should invest in one of those Gorgon Stare surveillance drones. Lift this fog of war.

More ›