Its only a problem if you want to run AI. If you don’t want AI locally or cloud based, then no need to spend the money on the high end 32GB model (for AI purposes) or paying for a cloud subscription. No one is required to get the 32GB model if they don’t want it.
Lummis, 71, told Capitol Hill reporter Pablo Manriquez on Monday, Feb. 9, that while she previously had said, “I don’t care” about Congress’ efforts to force the release of Epstein evidence, she now believes the issue was “worth investigating.”
The cynic in me thinks she ignored pages and pages of trump named referenced crimes and saw Clinton’s name at least once, so now she’s wants to make political hay only against Clinton.
yea, i’m surprised, 32GB is goddamn ridiculous for anything, let alone for a shitty hp branded autocorrect
32GB is actually considered the bare minimum for most of the common locally run LLM models. Most folks don’t run a locally run LLM. They use a cloud service, so they don’t need a huge pile of RAM locally. However, more privacy focused or heavy users with cost concerns might choose to run an LLM locally so they’re not paying per token. With regards to locally run LLMs, this would be comparable to renting car when you need it vs buying one outright. If you only need a car once a year, renting is clearly the better choice. If you’re driving to work everyday then clearly buying the car yourself is a better deal overall.
You are perfectly fine not liking AI, but you’re also out-of-touch if you think 32GB is too big for anything. Lots of other use cases need 32GB or more and have nothing to do with AI.
I agree with your frustration with subscription laptops. I hope people don’t use it.
I can almost guarantee you that nobody under the age of 30 gives a singly flying fuck about having an antenna on a television. They’re probably watching more than half their media on their phone or tablet anyway.
…and…
if you asked a kid in their twenties if they even knew what an antenna for a television was they’d probably go “what the fuck are you even talking about?”
I’m not sure that’s accurate. Gen Z and younger are apparently re-embracing OTA TV.
"The study found that younger viewers now over-index on digital antenna usage compared to their older (50+ year-old) counterparts (23% and 15%, respectively). " source
I’m much older but OTA TV is still a nearly daily use in our house even if the same content is available on various streaming services. DVR means skipping commecials while also getting a much better image quality than highly compressed streaming.
We found OTA TV is a great compliment to streaming. There’s no need to pay a cable/satellite subscription, you don’t have to constantly worry about that bill going up year over year or a local channel being blacked out because of contract disputes. There’s no “service” to have to worry about going out.
What is surprising about this? LLMs are giant memory consumers.
For instance analogue IO would be very welcome for many purposes.
I have a handful of the original Raspberry Pi A and B models that have analog audio built in (with 3.5mm jack) and they’re still in use today because of that built in analog audio out. Also honorable mention to the original Pi Zero which had the logic (but not the header pins) for NTSC out. I have a couple of those in use too because of their NTSC out capabilities.
I’m not understanding why he would want the station to be renamed “Orange Turd Station”?
after decimating the indigenous americans that have been here more than 10k years.
No argument on the truthfulness of your statement, but I’m not sure what that has to do with the premise of society enforcing the thought that the rich are rich because of god.
If you’re looking for specific historical knowledge, as in citations, here you go:
These are exactly the type of thing I was looking for. Thank you for sharing them.
I have never intentionally put words in your mouth. The best I can figure after rereading our entire thread is that you’re jumping around on different points but giving no clues in the conversation you’re doing that. As in, I’m responding to one of your points, but you’re providing a rebuttal for a completely different point of your own.
In this conversation I’ve been trying to restate what I’m seeing as your interpretation in an attempt to confirm we’re communicating, but then I get another response indicating we’re not communicating.
There’s two possibilities I see as to whats happening here:
OR
For the purposes of civility, I’m not going to make a judgment one which one these it is. I’ll let you give your downvote button a rest and simply bow out talking more with you today. Maybe in the future we’ll have better luck with one another.
I’m not looking for pedantry. I’m looking for clarity. You eluded to a specific action by robber barons in the 1900s. I’m looking for what that is because I’m seeing that idea predate them.
Robot automation has not lowered the quality of a Ford vehicle
I never said that and the quality of a ford truck is irrelevant to the assembly worker who lost their job due to automation.
You need to back up because you have gone down a tangent alone.
I agree we’re down a tangent, but I’m following the logic of your responses. This is a response to your original thesis: “AI robots can be utter shit”. Then you introduced the ford example for automation, which isn’t shit for assembly.
Which point to you want to back up to that would change our conversation path?
The notion that people won’t eat sawdust bread is demonstrably false with many historical examples proving you wrong.
I’m glad you saw those. I specifically chose sawdust in my example because of those events in history. Those support what I’m talking about. When the adulteration of the food became bad enough, people stopped eating it.
Your stipulation about zero flour is a moving goalpost and a strawman fyi
My “zero flour” comment is a response to your original thesis where you said: “quality of service can drop indefinitely.”
It can’t be indefinitely. There’s a point where people will stop consuming it when it gets bad enough.
Early colonized America used slave labor by racist christians. Those racist christians said they were supposed to be rich because god made them that way. That predates the robber barons of the early 1900s.
And yet youtube is still the dominant video host.
Youtube hasn’t descended to being unusable yet.
You’re missing the point entirely. If instead of luxuries you look through the lens of necessities perhaps you’ll see. Like replace cookies with bread and try tell me people will choose to starve first. Like obviously not.
I think you’re missing the point. If we substitute bread in the example I gave and they’re putting sawdust in it, then yes people will not buy bread made with zero flour, but instead made with sawdust. Yes, people will stop buying bread in that situation because they would die anyway because the bread doesn’t produce nutritional value.
Ask a ford employee 30 years ago about robot automation. Like this is not a new thing in the 2020s. The rich have a playbook for this.
Now you’re speaking against your original point. Robot automation has not lowered the quality of a Ford vehicle. If anything it has increased it. A robot can have assembly tolerances much tighter than a human. Where is the lowering of quality from a robot making the vehicle that your original thesis demands?
The short version is that the ultra wealthy were pissed about the New Deal, so they used fundamentalist Christianity to tie the idea of wealth to holy favor from Yahweh.
That concept existed WAY before the United State did.
The old idea was kings were rich because they were ordained to be kings by god. Questioning why the king was rich was questioning the word of god and punishable by death.
AI robots can be utter shit and they will still be leaps and bounds more efficient than the task specific automation that has been replacing human workers for decades.
I disagree with this, and we already have live examples today that are good analogs. Youtube is being flooded with AI generated slop. AI generated scripts, read by AI generated voices, over top of AI generated images.
I never seek these out, and actively avoid them when I can tell what they are before clicking on them. In that first 2 seconds of AI generated voice, I can tell this is slop and stop watching it seeking a human generated video instead.
As long as the rich maintain their monopolies quality of service can drop indefinitely. Doesn’t matter if AI robots suck ass when no human employed company can compete and every other option is just as ass.
It can’t. At some point the quality of the product drops to a level it is no longer a product. Lets say we’re in your theoretical dystopian future where the monopoly exists for cookies. There is no other place to buy cookies except from the monopoly. You posit that quality can drop indefinitely as there is zero alternative sources for cookies. So lets say the monopoly cookie brand was deciding to substitute some of the wheat flower with sawdust as a cost saving measure with the consequence being yet lower quality cookies. At a tiny fraction of sawdust you may notice it, but the sawdust cookie may still be better than no cookie. The monopoly continues to increase the sawdust content until the cookie contains zero wheat flour and is entirely substituted with sawdust. I believe even you would concede you would no longer buy the sawdust cookies at this point. Further, you would have stopped buying them earlier when the sawdust content became so high that the cookie was inedible to you even though it contained some wheat flour at that point.
This same thing will apply to Youtube. If the only thing left to watch on youtube is AI slop because no human creators exist, then there is no point in watching youtube anymore.
The point here, is that even with a monopoly on a product, as soon as the quality drops below a certain threshold (and this point is different for every consumer), the product stops being a product to them.
Not defending Musk, but the point of humanoid robots is to perform a job currently done by a human worker without modifying the process or tools. Dedicated robot arms are fantastic for factory work, but the jobs they do have to specifically be designed to be done by a robot arm.
As an example, you can’t put a robot arm at a human workspace and have it open a plastic bag, put an item inside it, and pick up a tape gun seal it with tape. For a robot arm to do that, the entire workspace, and extra robots would have to be added and programmed to accomplish the same task.
xAI must not have been getting enough investor interest and rather than admit it’s a stinker, he’s shackling spaceX with it.
This is the same thing Musk did saddling Tesla (and Tesla public shareholders) with the debt of the failed company Solar City run by his cousin.
Under which rock does he live?
Under the rock where reading comprehension exists apparently.
Where he was prompting for “What is 2+2?” to the LLMs, the accuracy of the answer was immaterial. At that step he was comparing two systems and simply needed a static question to give both system to compare the internal processes to determine why they arrived at different outputs (or a what appeared to be race condition/infinite loop for one) when the result should be identical to both irrespective of how right or wrong the answer is to the prompt. The LLM answer from the LLM could have been “ham sandwich” and it still would have served his purposes.
This is a pretty decent article and answered some of the exact questions I had when I heard about the recovered video.
That’s not what “end-to-end encryption” means. End-to-end encryption means only the sender and receiver have the ability to decrypt the message. The definition the author provided would be a match for “Zero-Knowledge Encryption” instead.