Good points. What’s novel information vs. wrong information? (And subtly wrong is harder to understand than very wrong)
At some point it’s hitting a user who is giving feedback, but I imagine data lineage once it gets to the end user its tricky to understand.
If I’m creating a corpus for an LLM to consume, I feel like I would probably create some data source quality score and drop anything that makes my model worse.
The bans are for under 16s, not just 7 year olds. Parents don’t control all internet activity for 15 years, at that age they are going to have some autonomy outside of the house.
I’m not sure there is a direct irl analog when it comes to controlling digital spaces, since they are personal by nature. and I think this is where the debate comes in.
Should parents be following their teenage child into every store to make sure they aren’t buying alcohol?
I get the concern with providing social media companies a government ID, I certainly would never give them one! I would just not use them. But they provide net negative value in my opinion so no loss.
I like the idea of FOSS parental controls.
Not to say it’s never the parents fault…
There is precedence though. We age gate: nicotine, alcohol, gambling etc…
we shouldnt expect parents to be monitoring children 24/7. actually, as children get older they should be given freedoms, parents have the right to expect our society has some guardrails.
Eh, don’t be rude. You are likely thinking of single gene mutations or other clear well defined problems.
My mind was more on polygenic diseases or genes with variable expressiveness. Where humans being humans we target things where we don’t completely understand the outcomes.
We screen for chromosomal abnormalities I don’t have a problem with that for example.
There is a strong possibility we would also get it wrong. Diversity is a strength. Who knows what tomorrow brings.
It makes me a bit sad that there is a whole article on a (very likely) mirage
500 million was specific to Claude Code, they are at 5 billion annual run rate and growing
I mean, I agree that a lot of money was spent training some of these models - and I personally wouldn’t invest in an ai based company. The economics dont make sense.
However, worst case, self hosted open source models have got pretty good, and I find it unlikely that progress will simply stop. Diminishing returns from scaling data yes, but there will still be optimizations all through the pipeline.
That is to say, LLMs will continue to have utility regardless if Open AI and Anthropic are around long term.
Some people are finding value in LLMs, that doesn’t mean LLMs are great at everything.
Some people have work to do, and this is a tool that helps them do their work.
The ‘cloud’ was a pretty big thing though… everyone used to self host, now only some self host.
AWS, GCP, Azure make a lot of money
Yeah exactly!
Could this be attributed to the driver mix changing?
It’s quite possible tesla drivers are worse in 2025 than 2024
I might be misunderstanding, but how is using a meme template, art?
Memes are lame. Generative images isn’t art. Both have been used in expressive ways in some cases, but generally it’s still shit internet culture.
Memes are worse than generated images in my opinion. Memes were the beginning of the end.
deleted by creator
I’ve never seen an LLM response be cavalier with credentials.
Yeah, after reading a bit into it. It seems like most of the work is up front, pre filtering and classifying before it hits the model, to your point the model training part is expensive…
I think broadly though, the idea that they are just including the kitchen sink into the models without any consideration of source quality isn’t true