Skip Navigation

User banner
smoothbrain coldtakes
smoothbrain coldtakes @ canis_majoris @lemmy.ca
Posts
0
Comments
527
Joined
2 yr. ago

  • I've seen this in a number of threads but I've never seen direct proof?

  • Apparently he's been integral in the whole growth process, I had heard that he personally interviewed most folks that are currently part of the team.

  • So, I came here with a bit of knowledge in Linux having fucked around with Ubuntu and Arch here and there, and I can tell you, even with a sturdy and non-rolling release like Fedora Silverblue, there are easily things I can do in Windows that just work without any additional overhead or configuration that simply does not work in Linux, like fingerprint sensors.

    You guys all say Windows sucks and Linux is the greatest thing since sliced bread but it still can't do the fucking basics that Windows does in spades. When I install Windows on a machine, I have nearly a 100% guarantee that every single component is going to work properly with minimal config. When I try to do the same thing in Linux, it's hours on the Arch wiki or deep into forums trying to figure out how to get something as basic as a fingerprint sensor to work. That's not convenient for the average user, and you guys are not the goddamn average user, because you are okay with shit not working out of the box and doing configurations for a lot of little things that you would otherwise just take entirely for granted as simply working on Windows.

  • Stuff without the guardrails, stuff that's been designed to produce porn, or totally answer truthfully to queries such as "how do I build a bomb" or "how do I make napalm" which are common tests to see how jailbroken any LLM is. When you feed something the entire internet, or even subsections of the internet, it tends to find both legal and illegal information. Also the ones designed to generate porn have gone beyond that boring shitty AI art style and now people are generating human being deepfakes, and it's become a common tactic to spam places with artificial CSAM to cause problems with services. It's been a recent and long-standing issue with Lemmy - people like Exploding Heads or Hexbear will get defederated and then out of retaliation will spam the servers that defederated from them with said artificial CSAM.

    I like copilot but that's because I'm fine with the guardrails and I'm not trying to make it do anything out of its general scope. I also like how it's covered by an enterprise privacy agreement which was a huge issue with people using ChatGPT and feeding it all kinds of private info.

  • Rogers pushed a BGP update that knocked a third of Canada offline. Did anybody step down? No! We were all just told to fuck ourselves!

    It knocked most of our financial services offline and cost us literally billions of dollars.

    Imagine taking responsibility for fucking something up. Amazing.

  • I could see it working for a fighting game but it feels like they've whiffed them recently. Wasn't Multiversus supposed to be a pretty big game? I remember it lasting a few weeks of hype and then going into hibernation, something about an early access release and then a season 2? I don't remember honestly.

    Mortal Kombat easily fits into live service bullshit, sadly, with all the skins and cosmetics that could be applied to battlepasses instead of cosmetic stores.

    All the great WB games in the past have never had this kind of monetization. All the Arkham games, Harry Potter; it's kind of sad seeing them take this approach in the future but it's just a cash cow even with a few whales.

  • His return deal totally capsized, he's out as CEO still. The old CEO of Twitter, Emmet Shear, is now in charge.

  • It's already easy to self host and we've optimized LLMs to run locally on not much serious hardware after we've trained them; I have GPT4ALL set up on my machine and it runs everything locally with my processor, no GPU or anything. Some of those datasets are uncensored, and I've seen what Stable Diffusion can do for image generation.

    I tend to use the GPT-4 built into Edge with my O365 corporate plan, because it suits my needs better for day-to-day challenges. It can still audit code and summarize things, which is all I really need it to do here and there.

  • Nothing about this is safe. It's easily the worst misinformation tool in decades. I've used it to help me at work, GPT-4 is built into O365 corp plans, but all the jailbroken shit scares the hell out of me.

    Between making propaganda and deepfakes this shit is already way out of hand.

  • They pushed it back. They've done so several times with Manifest V3.

  • Does it also mention that he basically deployed it exclusively in the third world, in places that the regulations were/are lax?

  • Crypto bro, wants to scan everybody's eyes with AI, has this weird startup that puts eye scanners in orbs and places them around the world where people are curious and stare into them having their eyes scanned.

    He's just a weird dude in general.

  • We've had some crazy things happen when you make a large map with ~10 people coming in and out.

    One of our friends focused on bus service in these few towns, set up a bunch of lines, and then logged off for a few days - his area grew to become basically four or five contiguous cities, it was pretty rad.

    Our endgame goal was to build maglevs hitting every single player's main city, and it took a while, but it was really fun when we finished it.

  • OpenTTD is great with a bunch of mods or a bunch of rules. My friends and I used to do yearly games and we outright banned air travel because it's just too overpowered and makes you have endless money.

    Trains though, thems the ticket. Really satisfying.

  • When I knew it was Mike McMahan who ran the TNG S8 Twitter account, I had total faith. He's a true fan, he wrote the show with reverence for the material despite it being jocular in tone, which is impressive and what made me fall in love with the show.

  • He was literally schizophrenic.

  • Sorry, I am running the 3090 on a rig running an Intel 12-7000KF. I haven't had much experience with Optimus configuration, since I don't think the 1650 has Optimus either.

    I haven't gotten Starfield to run on the 1650, but that was a spec issue. What kind of card are you trying to run it with? Might just be that Starfield is too intensive to run, which was the case for me.