

And we never will, because of the fundamental ways the technology works. LLMs needs nondeterminism in order to function. It’s truly a black box, which makes it a nonstarter in any discipline that demands accuracy and integrity and repeatability.
People keep asking me, and I haven’t really had an answer, but now yeah, I’m thinking I’m back.


And we never will, because of the fundamental ways the technology works. LLMs needs nondeterminism in order to function. It’s truly a black box, which makes it a nonstarter in any discipline that demands accuracy and integrity and repeatability.


Would a pihole have any effect on ads served in this manner?
Billionaires.


It removes any creativity I have in the process. Complex decision making is no longer mine, the only say I have is Accept or Reject (and Roll Again). It’s soul crushing because entire process now has no soul.
And my coworkers who have bought in completely are now so lazy they don’t even call in to meetings anymore, they don’t answer questions, they let AI run the entire SDLC. When they get replaced by some MBA who knows how to hit the same buttons, they will be completely blindsided. And they will have nobody to blame but themselves.


I haven’t been following what these are or why they’re flawed. It looks to me like they’re taking Two Factor Auth, and removing one of the factors (the password). Now all you need is a device? If the device is lost or itself compromised, isn’t that still a single point of failure?


Any opportunity to get a repeat purchase out of them.


It’s really just one guy, brand new account, who posted like 5 times in this thread. I’ve already reported and moved on.


Stage 5: We pipe money directly out of your savings account for nothing. Thanks for doing business with us!


Check this video from Gamer’s Nexus. They don’t even need an ethernet cable, they can just install ads/malware directly through Windows Update.


Article explaining why archive.today may have been blocked: https://www.techspot.com/news/111296-wikipedia-may-remove-almost-700000-links-amid-archivetoday.html
Now I can’t speak for the whole fediverse obviously, but I think it’s generally accepted that these are positive things:
Tesseract’s biggest sins aren’t against 1 and 2, but 3 and 4. A default blocklist isn’t necessarily bad, though its contents are debatable, and it has to be user-editable. But the DDOS script (and what an artless one btw), and the lack of transparency is pretty bad.


Check his profile, as a tankie he has supreme authority on what goes on in other fediverse instances.


Thank you for the explanation.


Wouldn’t the total number affect accuracy of the rate? I think one chart in the article showed something like 700 proposals for low income areas and 100-200 proposals for high income areas. As N approaches zero, the rate of resistance or cancellation is a lot more sensitive to smaller numbers of events.


Ok, I’ve read the article and come back. I have the same question. The only statement that tangents my point is this one:
“The lowest-income, least-educated neighborhoods resist most, even among the low-income, low-degree areas facing proposals,” he adds.
But the variable the article cares most about is what happens to data centers that encounter community resistance, vs ones that don’t. Yes they categorize those 2 groups by income level, but there doesn’t seem to be a chart saying “data centers are 5 times more likely to be proposed in low income areas compared to high ones.”
Care to point me in the right direction?


Not to be a wet blanket, but might that be because data centers are being built in working class neighborhoods at a higher rate than wealthy ones?
Same people who read Snowcrash and thought the Metaverse was so cool.


It was located in the city in which I live
“Hot singles in your area!”


Regardless of the effect on circadian rhythm, I wonder if there’s any research on eye strain. I personally felt less strain when turning on dark mode or sepia filters, even in the evening.
Through the same process of non-determinism I just described. Lawyers asked ChatGPT to find cases so they could be cited in their arguments, and it gave them “sources” that were just bupkis.
It’s the same chat window, going to the same flawed machine. If you doubt the machine the first time, what makes you want to trust it the second or third time?
https://en.wikipedia.org/wiki/Mata_v._Avianca,_Inc.
https://www.bbc.com/news/world-us-canada-65735769