Unironically yes, sometimes. A lot of the best works which its training samples are based on cites the original poster's qualifications, and this filters into the model where asking for the right qualifications directly can influence it to rely more on high quality input samples when generating its response.
But it's still not perfect, obviously. It doesn't make it stop hallucinating.
Any LLM won't have the right architecture to implement that kind of math. They are built specifically to find patterns, even obscure ones, that nobody knows of. They could start flagging random shit indirectly associated with gender like relative timing between jobs or rate of promotions, etc, and you wouldn't even notice it's doing it
They've just released support for running and subscribing to 3rd party labeling services, I'm sure somebody's going to make a filter for that you can subscribe to
Blog pingbacks have existed forever but for different purposes
Somebody has actually implemented a Bluesky commenting system too already
It's not even properly impermanent because somebody else can just archive everything, ironically giving them better overview than the actual members themselves
It's not outdated at all, but you need more words.
See diceware, 7 to 8 words is typically all you need
And Motorola had true wireless earbuds earlier, etc.
Apple is about polish, not novelty, but a ton of people are obsessed with the idea of Apple as being "groundbreakers" everywhere.
To support MU-MIMO / beamforming (multipath signals for multiple devices) they could also just add more flat surfaces inside the ceilings to make radio reflections/echoes less complex so that the signal processing doesn't get overwhelmed when the source is some distance away.
Plain absorbing material removes interference but doesn't let you use MIMO tech as effectively, because the newer higher end routers can use those reflections to boost the signal
I'm pretty sure the problem is the shape and reflections. This type of design creates echoes from many directions which makes it harder to pick up the signal at a distance
Because if the community solutions are good enough then half the articles about the shutdown will mention it
Training from scratch and retraining is expensive. Also, they want to avoid training on ML outputs as samples, they want primarily human made works as samples, and after the initial public release of LLMs it has become harder to create large datasets without ML stuff in them