Rant. You may come across articles like this, or speeches where someone ponders "how can this happen in the richest country in the world?" Well, we're only the richest country in the world because our wealth is fake – using money borrowed while increasing the National Debt that will probably never have to be repaid. It's not because the underlying economic principles are sound. Congress votes on a lot of stupid stuff that doesn't make sense, and borrows more money to pay for it. And problems arise, and someone says the absurd trope again.
––– 凄い –––
Interesting. FedEx almost went bankrupt, and was saved by a lucky session of gambling at the blackjack table by the CEO. And apparently something similar happened with Peter Lynch of Fidelity fame.
––– 凄い –––
11 predictions of how AI will transform the world. The common theme is that it's not the AI itself we need to worry about, but the people who can wield it. AI will allow many people to unleash their creativity and make things happen that were impossible five years ago. Those who can use AI will do better than those who can't.
––– 凄い –––
Confusion regarding Biden's forgiving of student loan debt. Biden wants to claim that the loan is forgiven, but it's not that simple. Students still need to pay. I am against transferrring the debt to the rest of us. But lately, that crazy idea has gained traction, and students with worthless degrees think they need to be made whole. This whole mess, or course, was another debacle started by Obama.
––– 凄い –––
This is why we need a platform like X/Twitter:
And this is precisely why the mainstream media wants people to leave the platform.
––– 凄い –––
Oregon's economic picture seems to be okay, but dig deeper, and you'll see that it's not strong at all. We just haven't had as farther to fall, like San Jose.
––– 凄い –––
70-year old man graduates med school. Now there's internship and residency – at least another year, probably more. Unless he just wanted the degree for, you know, just because.
––– 凄い –––
Thank goodness. Pediatricians aren't giving COVID vaccines because it doesn't make financial sense anymore. Declining family interest, too. So was all the expense and hysteria worth it?
––– 凄い –––
For radiology applications, GPT-4 with Vision faired poorly, getting only 47.8% correct on image problems, even though it got 81.5% correct on text-based questions. Paper here. It appears from the Methods section that the researchers did not do any fine-tuning but used the model straight out of the box. Fine-tuning would likely improve performance.
––– 凄い –––