stri8ed

6 ms·

1.
by stri8ed 3mo ago
It's not a conspiracy. There's a finite amount of compute available, and they will sell it to the highest bidder. If another company can produce the same intelligence for cheaper, then they will drive the price down.
2.
by stri8ed 4mo ago
Output cost is 3x from Gemini 3 flash.
3.
by stri8ed 4mo ago
Given the cost increase associated with this model, and previous model releases, I think the size is trending upwards, not down.
4.
by stri8ed 11mo ago
Is it fair to characterize a founder of a failed startup as a con man? I.e. did he make claims that were factually untrue, and intentionally deceived investors?
5.
by stri8ed 1y ago
What is the counterfactual? Without knowing the number of attacks prevented by these tools, we don't know what the baseline would be.
6.
by stri8ed 1y ago
Having adequate law enforcement training and funding, is not mutually exclusive with leveraging technology for more effective enforcement. In fact that's where some of the funding goes. I would be interested in seeing some data reflect
7.
by stri8ed 1y ago
Not a chance. Even if American companies did abide by it, there is no reason Chinese companies would. And good luck definitely proving that a model trained on it.
8.
by stri8ed 1y ago
That problem along with its many solutions are surely littered throughout the training data. Not to mention, it would be trivial to overfit on that problem. I don't know why people still reference that.
9.
by stri8ed 1y ago
It's a result of the system prompt, not the base model itself. Arguably, this just demonstrates that the model is very steerable, which is a good thing.
10.
by stri8ed 2y ago
Trained at least in part on Chat-GPT data.
11.
by stri8ed 2y ago
It would happen in China regardless what is done here. Removing billionaires does not fix this. The ship has sailed.
12.
by stri8ed 2y ago
The benchmarks agree as well.
13.
by stri8ed 2y ago
Isn't that how previous models were, before the attention is all you need paper?
14.
by stri8ed 2y ago
Or be in the business of building infrastructure for AI inference.
15.
by stri8ed 3y ago
You can modify this, by setting the GPT4 system prompt, with instructions of your preferred response style.
16.
by stri8ed 3y ago
For programming, GPT4+. I was excited to switch to Claude after hearing all the positive anecdotes. Having tried it, I'm very unimpressed. It spouted complete, confident sounding nonsense, when I prompted it with a bug I was trying to
17.
by stri8ed 3y ago
It's impossible to prove.
18.
by stri8ed 3y ago
I don't think you can infer someone's age and skin colon from that post.
19.
by stri8ed 3y ago
In humans, a software problem can become a hardware problem, due to plasticity in the hardware. See alcoholism for example.
20.
by stri8ed 3y ago
Do existing LLM's not already train on this data?
21.
by stri8ed 3y ago
In such a scheme, wouldn't synonyms of the same word be no closer to each other, than any other random string?
22.
by stri8ed 3y ago
What is the rate of increase, and how does it compare to CPI inflation?
23.
by stri8ed 3y ago
How does the quantization happen? Are the weights preprocessed before loading the model?
24.
by stri8ed 3y ago
Let's not forget Google maps, which already integrates other ride share platforms.
25.
by stri8ed 3y ago
Very important distinction here, and that is self volition. To be clear I'm not saying it's good or bad. I think it's a genuine question. What if there are simply not enough jobs available at the rates that people want to be
26.
by stri8ed 3y ago
Would there be any demand for those gig works, if it had to pay comparable to full employment? And would that be a good thing?
27.
by stri8ed 3y ago
It's the cost factor. Using AI, it's economically feasible to contact a million people, even if only a fraction bite. Not unlike the current spam calls.
28.
by stri8ed 3y ago
The authors address that by looking at hospitalizations for self harm.
29.
by stri8ed 3y ago
In the article, the author makes a point to include objective metrics like rate of hospitalizations, and the data is quite conclusive in showing an actual increase. So these issues definitely did not exist before, at the same scale.
30.
by stri8ed 4y ago
For the same reason that people oppose monopolies. Unions are effectively a monopoly on the labor market.

More ›