Browse Comments

The goal isn’t to use their likeness. The goal is to train an AI using a data set to be able to generate artificial humans as needed. Their actual face would never appear. It is going to happen. You don’t need even need actors, and some people will agree to be scanned.

Jesus Christ, it isn’t about whether or not AI writes better than a fucking human you idiots. Businesses need to allow HUMANS to make good livings. WE CANNOT support studios that will cut human jobs for AI to save a buck. Can’t do it. We will of course, but we shouldn’t. We are the problem. Too fucking many of us are too lazy and apathetic to actually sacrifice enjoyment or convenience …

There are so many articles about the writers strike against AI. That’s not what they’re striking over. It’s considered a minor issue. The strike is about residuals. It’s about studio notes and producers that found a loophole to keep writers on staff longer for the same total wage. It’s about decimation of writers rooms, and forcing changes to creative processes to cut costs. The AI is like a ch…

I think one of the core points of the latest Writer’s strike was AI protection. Actors just need to do the same. Basically entire industry needs to get AI protection clauses into their contracts. AI can help, patch some rough spots, but humans need to be at the foundation of it all.

Plus literally the only reason an AI would know to use that term is because it’s been in use for hundreds of years and is used in the training data in the first place. These people have zero concept on how LLMs work. I’d bet if you asked the person to theorise how LLMs work internal in even the most general sense, they’d probably say it was a “system of heuristics” or a bunch of if-then statement…

Facebook actually has really good AI tech and surprisingly generous with their open source contributions. They are the ones that released the llama model.

AI will move so fast it will either save us or destroy us before climate change. Maybe both.

It was already happening. The only thing AI does is speed it up a little.

If your job can be so easily replaced by AI, it will. I fear for all the email summarizers and 3-day trip planners to excessively traveled cities.

>There’s more than enough jobs at SMEs to go around, but young Koreans would rather be unemployed than work those jobs, lest be seen as “not keeping up”. This sounds pretty similar to the Rust Belt. A disproportionate number of men who had good manufacturing jobs were unwilling to take jobs that paid less, so they just stayed unemployed forever.

The guy claimed to be an artist but that feels bunk to me. Had the vandal not been a serial sculpture smasher, this could’ve been an interesting artistic statement. One of Ai Wei Wei’s most well-known works, [*Dropping a Han Dynasty Urn*](https://www.guggenheim-bilbao.eus/en/learn/schools/teachers-guides/ai-weiwei-dropping-han-dynasty-urn-1995), involved smashing 2,000 year old urns as a critique…

Literally garbage. 99.9% of what ai will be used for is to produce crap because people are idiots. Internet will become irrelevant. No one will be able to trust anything.

This is a really interesting threat. I’ll be curious to see what happens over the next ten years. But my hypothesis is that AI, in particular, will level-up both the capabilities and expectations of a junior dev. So then they take on what used to be junior+ or mid-level work.

So this is the reason why Nvidia banned GPU drivers updates to Russia this week

Of all the blatantly idiotic/traitorous ideas I've heard so far (depending on which end you are) this one has to be up there within the top 10.

Step 1. Go to https://ollama.com/ and download it Step 2. Open the terminal on your Mac or whatever Step 3. Type `ollama run deepseek-r1` and hit enter Step 4. Enjoy Optional Step 5. - if you want a ChatGPT style UI instead of the basic terminal, install this - https://github.com/open-webui/open-webui It's absolutely fucking mind blowing that we can run these LLMs locally without any intern…

Don’t get me wrong, LLMs definitely help especially with boilerplate code (I myself hate writing E2E tests manually) but I have to check every single output for fundamental mistakes

Those layoffs were not actually related to development efficiency. People on my LinkedIn from all types of roles were cut and AI isn’t doing their job nor supporting it

It’d be better if the CEOs were replaced by AI. Why go for the small job that pay less money when you can go for the big jobs that pay most of it.

I feel like you're projecting some things, the post you're responding to didn't have an angry tone and didn't imply any kind of abuse. I lso think that you have some misunderstandings about how Chatgpt works. It's an advanced hat engine, buy it is not intelligent. Anything that it can teach you may be right, or it can be disastrously wrong. I don't think that it's a good idea to think of Chatgpt…