Skip Navigation

User banner
Posts
8
Comments
156
Joined
2 yr. ago

  • I've used ESET NOD32 for a long time. It costs money, but it's relatively light weight, doesn't get in the way of any of the gaming or video editing or programming work I do, and it has smacked my hand a few times clicking on risky links, blocking a few downloads (one or two, ever) and often blocking scam websites that attempt to run javascript crypto miners. Your experience may vary if you're planning on using "rescued" media or applications. I don't have have the full internet/devices subscription, just the one for personal computers that does the actual virus detection.

  • This is how I imagine China and North Korea in international relations.

  • "What if we just call it the other thing? The one people like?"

    "... my god."

  • Don't worry, its just a concept design from Carota. Zero chance this gets out the door.

  • I had to massage the prompt and have it rewrite it three times, but it finally nailed it. The dream journal cracked me up.

  • Personally I don't think copyright holders really have a leg to stand on as far as that goes. Simply having and using a copyrighted work isn't a violation, and the work that is produced in the form of a trained neural network is the very definition of transformative. I also think Meta would have the same issue with trying to use a copyright claim for someone using their llama output to improve other non-llama models. That's why they had to slip it into a terms of service.

    I guess what you might see going forward is every book that's published comes with a user agreement you agree to by opening the book... But that doesn't sound practical in any sense.

  • That wouldn't even bother me so much if they weren't trying to lock people into the "llama ecosystem" by restricting its use to only improving other llama models.

  • Sometimes ChatGPT can be legitimately funny. I asked it to fill out the permissions application as Meta.

    A high step count might be rewarded with an ad for comfortable running shoes; prolonged inactivity could trigger promotions for ergonomic chairs.

    LOL


    Dear App Store Reviewer,

    We're pleased to outline our reasons for the various permissions required by our Threads application.

    Permission #1: Location Access to location data allows us to provide a more personalized user experience. Localized content, such as debates about neighborhood parking restrictions, would facilitate more engaging interactions. It also enables more accurate restaurant ads.

    Permission #2: Health Data With health data, we can offer timely suggestions for wellness activities based on user habits. A high step count might be rewarded with an ad for comfortable running shoes; prolonged inactivity could trigger promotions for ergonomic chairs.

    Permission #3: Microphone Access Audio transcriptions serve to identify users' conversation topics, allowing us to tailor content to their interests. Casual mentions of Italian cuisine could prompt an influx of pasta maker ads, enhancing their culinary explorations.

    Permission #4: Camera Access Camera access improves our ability to understand user preferences through their captured images. Frequent snapshots of their pet parrot might lead to a subtle increase in bird seed and cage cleaner advertisements.

    Permission #5: Contact List Contact lists assist us in identifying common interests within a user's network. Frequent discussions about vintage car restoration with certain contacts might result in targeted ads for classic auto parts, supporting their hobby.

    Permission #6: Calendar Calendar sync provides us insight into users' upcoming events, allowing for timely content suggestions. If a user's calendar reveals a dental appointment, we might find it apt to provide ads for mouthwash and high-end toothbrushes.

    Permission #7: Biometric Data Biometric data offers enhanced user security, while also informing us about their stress levels. Irregular heartbeats during work hours might indicate a need for calming ads featuring spa retreats or meditation apps.

    Permission #8: Browsing History By understanding a user's browsing habits, we can refine the content we present to them. Browsing for second-hand furniture might result in a myriad of ads for antique shops, aiding in their quest for the perfect vintage armoire.

    Permission #9: Social Media Account Linking Linking social media accounts allows us to understand a user's broader digital footprint, leading to a comprehensive content curation strategy. Regular 'likes' on pictures of home-baked bread across platforms might trigger an increase in ads for artisanal bread making kits.

    Permission #10: Dream Journal Entries Our most innovative request, access to users' dream journal entries, could provide a unique perspective on their subconscious interests. Dreams of tropical beaches could result in a range of ads for sun hats, SPF creams, and all-inclusive resort packages.

    We trust these explanations clarify our intent: providing a user-centric application experience, catered to individual lifestyles, with the necessary bonus of extremely personalized advertising.

    Yours sincerely, The Facebook Team

  • I thought the point was a mental BDSM exercise where you come to others for help and are instead punished for your ignorance.

  • Posters aren't saying that its impossible to put search results through an LLM and ask it to cite the source it reads. They're saying that the neural networks, as used today in LLMs, do not store token attribution in the vocabulary or per node. You can implement a system for the neural network to work in that provides it the proper input (search results) and prodding (a prompt that encourages the network to biasing toward citation), not that the single LLM can conceptualize of that on its own.

  • If there were ever a man that didn't deserve any sympathy for something like this, it is Mitch McConnell.

  • Good point. I think I'll do whatever I want with it and just keep my trap shut. Good luck proving anything Zuck.

  • I have a feeling that this is going to go similarly to Stable Diffusion's big 2.0 flop. SD put its limits in through training data. Meta put in its limits via terms and conditions. The end result for both will still be that the community gravitates toward what is usable with the most freedom attached to it. The most annoying part of the TOS is that you can't use the output to improve other models.

    Fuck you Meta, I wanna make a zillion baby specialist models.

  • Consider switching now. Your continued usage is contributing to the user numbers they see reported that gives them the confidence to pull this shit in the first place.

  • Hey.

  • I don't have a quality comment, just wanted to say I fuckin' love LibreOffice. Have used it or its predecessor since I was in high school at the turn of the century. Lord that makes me sound old. I think LibreOffice actually forked when I was in college, so mid 2000's.

    Anyway, great software. Highly recommend. 10/10.