Yeah. Kind of crazy I can lose access to a 20 year old acct and Google is just like 'meh can't verify you' and we all just think that makes sense. I should be able to somehow verify who I am and get reinstated.
Honestly, I think some sort of blind verification system is what the internet needs.
Some kind of notary-like system where a verification happens locally with a person, and then no info on the person gets passed out except the verification to start an account.
Depends on the size of the agency. You can already guess a SSN based on the range of numbers used. If you were targeting the youngest or oldest person at a small agency you could probably get a high percent chance of getting a match.
Put the money towards some purpose that you think could really benefit from it. Giving it to ubisoft or silksong is, comparatively, useless. Giving it to a true indie dev that hasn't hit success, and giving a thorough review, could actually make a big difference.
Leafs, at least historically, have been air cooled. Fine if you're in the northwest, but really bad for battery health if you're somewhere like Arizona.Battery health can be tricky to verify. Don't just count on the bars.
If you're somewhere cold, anticipate that 150 mile range to cut in half with the heater going.
Streisand didn't want aerial images of her house to be available on the internet. The subsequent outrage made it so those pictures got on newspapers nationwide.
I started saving what I put into terminal into a Google doc. That way if something gets screwed I have a record for how to undo it.
Also, don't copy paste from chatgpt to terminal. And make sure the source you are using isn't 7 years old.
Cgp grey has a video about algorithms that is very accessible, and gives a good visual display of how they work. It's kind of stuck in my brain and I think back to it every now and then. I think the same explanation will apply to LLMs.
To sum up - they are good at making associations between data - but nobody knows which linkages are made or how/why the final results of the LLM are made. That's why you get outcomes like chatgpt making statements with racist undertones - if there is racism in the training data (e.g., racist internet comments) those associations will carry thru to the final result.
Thank you. Fedia puts the cursor between the brackets, which makes me think the URL goes between the bracket