Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
skeptic_ai
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
11 ms
·
91.
▲
by
skeptic_ai
4mo ago
I’msure they will just ensure they will never reach 100m and stash their cash somewhere hidden. So won’t work that well. But might work for a one off cut. Which is good.
92.
▲
by
skeptic_ai
4mo ago
I don’t understand how I grew up thinking USA is the gold standard is good and China just make cheap copies and is bad. But these news really changes my view on China and USA. I can’t believe it almost.
93.
▲
by
skeptic_ai
4mo ago
Why China wouldn’t release?
94.
▲
by
skeptic_ai
4mo ago
When you fire the bottom x % you have incentives to kneecap your “colleagues” to preserve your own job
95.
▲
by
skeptic_ai
4mo ago
With ai I can get government bodies to do their job. Escalate and make formal complaints until it’s done. Manually would take ages to find the right law and draft a proper complaint.
96.
▲
by
skeptic_ai
4mo ago
What’s the point when they will remove those guardrails when competition reaches their levels. Shows that they don’t Reddit care about “safety” at all
97.
▲
by
skeptic_ai
4mo ago
Have you seen many corporations complaining and caping usage to 20-200usd per developer per month. I doubt will change much. Many are considering on premise now.
98.
▲
by
skeptic_ai
4mo ago
So if I make a script to spam everywhere with 0.1% odds that is fake I can’t be sued? Just because I spam millions of times per day means I shouldn’t be liable for what I do right? But someone saying one time should go to prison. Makes sens
99.
▲
by
skeptic_ai
4mo ago
That’s too easy. Would be nicer memory leaks, intertwined spaghetti code, time dead bombs, bugs based on time
100.
▲
by
skeptic_ai
4mo ago
Tbh I feel what you say it’s just to feel good about how bad our situation is. Oh, their situation is much worse, don’t worry about us, we are so much better.
101.
▲
by
skeptic_ai
4mo ago
Sp500 is not passive. Keeps updating the rules every few years. https://www.reddit.com/r/stocks/comments/1tvuhgy/sp500_100_y... See top relevant changes in 100 years
102.
▲
by
skeptic_ai
5mo ago
We all know they get paid by musk to load up on overvalued stocks so musk can get some cash from pension funds, the pay off a bit Russell’s for bending the rules. No one in their right mind would change rules to buy space x. What profit mus
103.
▲
by
skeptic_ai
5mo ago
Don’t forget context. Basically I have 2 billion input and 1 million output. Every prompt you do, sends back the whole thing again and again. Let’s say you have 500k context used, you send 10 messages is 5 million. 100 messages 50 million.
104.
▲
by
skeptic_ai
5mo ago
But you rent a studio and they have a 4 bed house. You compare apples and oranges
105.
▲
by
skeptic_ai
5mo ago
You can’t compare buy house with rent a flat. You need to compare same property style/type. Otherwise you just lying you yourself. My friend said: I can rent this 3 bed room for 6k. But if I buy a 1 bed I need to pay 2k. So it’s 3 time
106.
▲
by
skeptic_ai
5mo ago
Exactly lol
107.
▲
by
skeptic_ai
5mo ago
Npm install qr-made-up-name Can show qr in console. How do you stop that?
108.
▲
by
skeptic_ai
5mo ago
I really hope eu can extinguish Google before they extinguish all websites. Will be an exponential death of website very soon once they lose traffic.
109.
▲
by
skeptic_ai
5mo ago
For me down 99%
110.
▲
by
skeptic_ai
5mo ago
I did some monitoring. 15 accounts, 300 millions tokens input, 200k output went to 0 the 5h quota in 7 hours. 4 parallel tasks. I think 300 million is too low. For reference before I could do more than 1 billion on same conditions.
111.
▲
by
skeptic_ai
5mo ago
So if I exchange your Rolex with a fake one and then you try to sell after 3 years and you notice it’s fake, it’s fine for you?
112.
▲
by
skeptic_ai
5mo ago
I developed for 15 years. I don’t think I can do with AI anymore. Why would I even want to do that? It’s like telling a car driver to build an engine.
113.
▲
by
skeptic_ai
5mo ago
If you have checked your AI would have said is not the cert
114.
▲
by
skeptic_ai
5mo ago
They did it before AI.
115.
▲
by
skeptic_ai
5mo ago
If you’re so sure just make a leveraged bet and become a millionaire. Put your food where your mouth is if you’re so convinced
116.
▲
by
skeptic_ai
5mo ago
Grok auto: 1 “Strawberry” has only one “e”. S T R A W B E R R Y Gemini: There is *1* "e" in the word "strawberry". Seems fine
117.
▲
by
skeptic_ai
5mo ago
Your brain is also an autocomplete at this point. Notice how you write each word, one after the other, flawlessly
118.
▲
by
skeptic_ai
5mo ago
Porn or social media or fake news destroy kids brain more? I can’t even tell
119.
▲
by
skeptic_ai
5mo ago
I never touched kubernetes and in 1 week I have a few nodes running and i understand a lot of it. Not perfect but not bad.
120.
▲
by
skeptic_ai
5mo ago
But right now you also do the job of 10s compared to x years ago.
More ›