Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
rmbyrro
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
41 ms
·
691.
▲
by
rmbyrro
3y ago
There's a good chance that the fear of China taking over the AI space worldwide may end up being stronger than OpenAI's push for regulation. Politicians know the later is real, and they also know that the "Terminator" fe
692.
▲
by
rmbyrro
3y ago
They want monopoly. Or, worse case, olygopoly. Which leads to an unrivaled, long-term money making machine and power.
693.
▲
by
rmbyrro
3y ago
It's because Salt Man doesn't believe it. He knows it's not the end of anything, it's the start of an insanely lucrative market. What he really wants is to capture as much of this pie as he possibly can. In order to do t
694.
▲
by
rmbyrro
3y ago
That sounds to me like opening console and messing up with the website's JS code :D
695.
▲
by
rmbyrro
3y ago
Theranos was a terrible comparison
696.
▲
by
rmbyrro
3y ago
Perhaps intentionally misleading?
697.
▲
by
rmbyrro
3y ago
AI is taking work away from lawyers, and instantly creating more work for lawyers. Ain't that interesting to reflect upon? I speculate there is a hidden force in the universe, something physicists are yet to identify, which mandates: &
698.
▲
by
rmbyrro
3y ago
It gave me the impression this person is concerned about helping teachers navigate this new and challenging reality. Call me crazy, but that was the impression I had from the comment and the course website.
699.
▲
by
rmbyrro
3y ago
You might be influenced by the perception that LLMs are autonomous. They're just tools. The outcome depends heavily on the user skills. That's why prompt engineering is a thing. Heck, even Google results varies depending on the se
700.
▲
by
rmbyrro
3y ago
I found lots of good value in their publication as well. Especially for teachers, who I believe (most at least) have no clue about prompt engineering and how to talk to an LLM.
701.
▲
by
rmbyrro
3y ago
I thought it gives good guidance. Of course it's not a 4-hour in-person workshop, like what you're proposing. But it already adds positive value. It covers a good amount of the topics your course covers, I think. Introductory-leve
702.
▲
by
rmbyrro
3y ago
How does it change the interpretation of the sentence if you consider "their" to be user's hardware or Nvidia's hardware?
703.
▲
by
rmbyrro
3y ago
> do they still work here, what’s their email And what happens when you send the wrong message with sensitive data to the wrong Brian because of that?
704.
▲
by
rmbyrro
3y ago
Come'on, there's only one hardware maker there, no need to specify. I did provide an example of illicit usage, could you please just read it again?
705.
▲
by
rmbyrro
3y ago
> plenty of real world use cases Any examples come to mind? I can't see a case when data retrieved from a database will be wrong, someone will do something with it, and it's fine.
706.
▲
by
rmbyrro
3y ago
> killer use case for LLM I hope you're referring to 'someone can eventually get killed by this'. For the love of humanity, please don't use this in any real world case. Llama will generate mistaken queries in at leas
707.
▲
by
rmbyrro
3y ago
Just use "on-prem"
708.
▲
by
rmbyrro
3y ago
For a minute, I thought the title was about Linux helping Nvidia block users from illicit usage of their hardware (like using consumer GPUs on data centers). And was like: "WTF? No way, that must have been a Court order". And then
709.
▲
by
rmbyrro
3y ago
Really? Do you have any examples to share? I'd be surprised to see that in action.
710.
▲
by
rmbyrro
3y ago
All this discussion shows that the real value is in maintaining the software, not just the software itself. What these companies are doing is basically abandoning the open source code they developed with the support of a huge community. The
711.
▲
by
rmbyrro
3y ago
Mind if I ask how is this related to the topic?
712.
▲
by
rmbyrro
3y ago
If? https://pubmed.ncbi.nlm.nih.gov/14964355/ https://cbmjournal.biomedcentral.com/articles/10.1186/s13021... https://www.eea.europa.eu/highlights/big-potential-of-cuttin
713.
▲
by
rmbyrro
3y ago
Perhaps you mean "open source" doesn't mean anything to you. Which is fine, but still quite surprising to hear in a hacker forum in 2023...
714.
▲
by
rmbyrro
3y ago
"to CONSTRUCT the image" [in a way that would generate more buzz and press?] I couldn't avoid wondering...
715.
▲
by
rmbyrro
3y ago
The worry is that biodegradation generates greenhouse gases. So if we avoid coffee waste from biodegrading, we avoid bacteria from producing greenhouse gases. Might people be freaking out too much about this stuff?
716.
▲
by
rmbyrro
3y ago
I suppose the goal of your message was to assert the security of the country's voting system. I think it's naive to think that any digital system is secure. What we try to do is to make hacking not worthwhile by raising the diffic
717.
▲
by
rmbyrro
3y ago
Exactly. People are only trying to keep up with inflation created by easing monetary policy and supply-demand imbalances during COVID and post-COVID.
718.
▲
by
rmbyrro
3y ago
Most companies don't use Snowflake or Databricks
719.
▲
by
rmbyrro
3y ago
This tool can extract data in a structured format from virtually any website, with any HTML structure. With Beautiful Soup, you'd need to explicitly tell where each piece of data exists referencing HTML tags, ids, classes, etc. For eac
720.
▲
by
rmbyrro
3y ago
Have you just found another way to bypass GPT's "safety" measures?
More ›