thwayunion

9 ms·

31.
by thwayunion 4y ago
"seemingly innocuous" is an impersonal construction referring to the reader, but not any reader in particular. The Fed itself, at least some presidents thereof, was aware of this risk. Toward the end of the article: "When the
32.
by thwayunion 4y ago
AFAIU, Fidelity's cash deposits are FDIC insured but the default "core" position -- where uninvested money goes to sit -- is SPAXX, which is not FDIC insured. That position is mostly U.S. Government Repurchase Agreements, wit
33.
by thwayunion 4y ago
The part on education was the most cringe for me. As a general rule: when rich big egos try to improve education, it ends poorly. Consider the Gates Foundation itself trying to MBAify teaching and then being astounded when it (a) didn'
34.
by thwayunion 4y ago
Also as long as you don't mind below market rates and job security tied to federal budget allocations. Or your paycheck disappearing for an unknown period of time whenever there's divided government.
35.
by thwayunion 4y ago
> Lots of exams are designed to prove certain knowledge given safe assumptions of the known limitations of humans, which are completely wrong for machines. The relative difficulty of rote memorization versus having an accurate domain mo
36.
by thwayunion 4y ago
>> We already know this is about self-driving cars. Passing a driver's test was already possible in 2015 or so, but SDCs clearly aren't ready for L5 deployment even today. > Who told you that? Passing a driver's t
37.
by thwayunion 4y ago
> I'm assuming we will solve the hallucination problem It's unclear what this would even mean, since "hallucination" carries a surprising number of different definitions and commentators are rarely precise about wha
38.
by thwayunion 4y ago
More-over, in most states, passing that bar isn't enough to practice. In most states, the bar exam is just one component. Humans also need to pass several demanding courses in law school, and to get into law school they need to get a B
39.
by thwayunion 4y ago
Again, because machines have different failure modes than humans.
40.
by thwayunion 4y ago
Most professional mathematicians work at universities. Even at well-ranked R1s, professors spend a non-trivial amount of time teaching and advising. The ones who do spend a massive amount of time doing mathematics are the least likely to be
41.
by thwayunion 4y ago
A good benchmark provides a strong quantitative or qualitative signal that a model has a specific capability, or does not have a specific flaw, within a given operating domain. Each part of this difficult -- identifying/characterizing
42.
by thwayunion 4y ago
A good benchmark is not simply a set of unit tests. What you want in a benchmark is a set of things you can use to measure general improvement; doing better should decrease the propensity of a particular failure mode. Doing this in a way
43.
by thwayunion 4y ago
Absolutely correct. We already know this is about self-driving cars. Passing a driver's test was already possible in 2015 or so, but SDCs clearly aren't ready for L5 deployment even today. There are also a lot of excellent example
44.
by thwayunion 4y ago
Can't replace it. But with some trivial API glue and a cron job, you can get away with storing next to nothing in the bank account on most days as long as all of your credits and debits are scheduled and upper-bounded. E.g., if the deb
45.
by thwayunion 4y ago
I'm so confused by what role a CBDC is even supposed to play... why would the government establish a new currency pegged 1:1 to its existing currency? Why not just... allow normal Americans to keep deposits at federal reserve banks?
46.
by thwayunion 4y ago
> How do we get society to agree to funnel more resources into AI safety? Regulation and liability. > I don't think your viewpoint and the article's viewpoint (or at least Ding's viewpoint), despite its snappy title,
47.
by thwayunion 4y ago
People used to anthropomorphize Markov chains like this. I've been through this rodeo before, at least a few times. I did a little lab on GPT with middle schoolers as part of a Science enrichment activity. Without prompting, the entire
48.
by thwayunion 4y ago
> Doesn't this go hand-in-hand with slowing AI research down and focusing more on safety engineering and best practices? I don't see how this is given. It could very well be the opposite. Consider self driving, for example. B
49.
by thwayunion 4y ago
This is more a statement about how bad technology hype has gotten of late. Having "tangible applications" is a very low bar. If we paused for every technology that has "tangible applications" we'd probably be enteri
50.
by thwayunion 4y ago
Am I the only one who thinks the AI doomsayers are even more ridiculous than the AI cheerleaders? ChatGPT may be useful and so on, but dear god. It's still just a fucking chat bot. We need to regulate AI and get better at building safe
51.
by thwayunion 4y ago
> direct bets AGAINST climate models... None of the things you mention after this sentence have anything to do with climate models. > cenovus up, pembina up, taiwan semi up At best, this is equivalent to saying health nuts have be
52.
by thwayunion 4y ago
The OP was asking a different question with a different answer -- OP wanted to know if novel ChatGPT output is copyrightable, not whether existing copyrights might cover ChatGPT outputs. That said... > The compression ratio of the sou
53.
by thwayunion 4y ago
Modern markets are obviously legal abstractions. The tech labor market would not exist as it does today without substantial very specific regulations of various parties' behavior; e.g., stopping me from: 1. copying my employer's c
54.
by thwayunion 4y ago
> Parents in cities have been arrested for letting their 10 year old kids play in the public manicured park unsupervised. Google "Free Range Parenting" as a city thing. Okay. I googled it. I'm seeing articles about paren
55.
by thwayunion 4y ago
Right... that's why I am providing OP with the advice that many rural communities are religious communities, a fact of which he may not be aware. Or, if he is aware, may not have thought through the implications. We're all posting
56.
by thwayunion 4y ago
I am giving OP advice on how to succeed at moving out of a city, not attacking your lifestyle. You seem mostly ignorant about what city life is like, and therefore are unlikely to give OP good advice about the differences between urban and
57.
by thwayunion 4y ago
If climate models were predicting the temperature of your city on March 21, 2120 then your intuition here would be wise. But that is not at all what these models are predicting. They are predicting global average annual temperatures and oth
58.
by thwayunion 4y ago
This is it -- colleges and universities are desperate for butts in seats ~> admissions standards slip ~> first year is remedial ~> lots of stuff gets dropped in later years. Another major consideration: "CS degree" is a m
59.
by thwayunion 4y ago
> Friends within a couple miles Which may or may not exist within 2 miles. Or even 10. Again, you have to be more intentional. In a city you can buy pretty much anywhere and the kid will have neighborhood friends. When moving to a rura
60.
by thwayunion 4y ago
No. I'm thinking of rural areas. I know the difference... weird for you to assume I don't, honestly. > hunting supervised So not independent. At all. > fishing, riding their bike places, exploring or "hanging out&quo

More ›