Mostly nodding along, with a few of these aged in interesting ways from where I'm sitting.
The drones bit hurts the most. There's a war an hour from our border eating FPVs by the millions, and Poland - sitting on batteries, motors, chips, a generation of engineers - has not stood up a real domestic drone industry. Money is there. Will is there. We just... haven't shipped. That should keep ministers awake.
EVs are worse. Izera is a punchline at this point. Noah literally called the play in 2024 - "don't bet on one champion, run a bunch and let them fight" - and the state did the exact opposite. We picked one horse and it never left the stable.
The Korea idea, on the other hand, Noah might have undersold. Framework agreement is for ~1,000 K2 tanks. By 2030 Poland will field more main battle tanks than Germany, France, the UK and Italy combined.
Rest holds up. "Try all the things" is right - we're just very uneven at the trying. Defense procurement: shipping. Civilian industrial policy: not so much. Software still works the way it always has: quietly, in apartments, mostly without the state in the loop. Which honestly might be a feature.
I can't imagine doing something like this anytime soon, but I will probably follow the advice around having AI check the return against all my forms and previous years return before submitting.
I was explaining this to my elementary school aged kids just a few days ago. We were eating in a restaurant and I told them that when I was their age most restaurants had a smoking and non-smoking section. Of course the smoke did not respect the invisible barrier. The idea that people could just smoke indoors and it was normal really blew their minds.
High school boys bathroom was basically a de-facto smoking lounge. It was banned but kids still did it. They occasionally cracked down, but the smell was permanent.
There was also an unwritten understanding that it was preferred the boys went out back to a certain door to smoke outside there instead and wouldn't get in trouble if caught.
I went to bingo years ago and there was a glass partition between the smoking and non, but it didn't go to the ceiling. So you'd sit in the non and just watch a wave of cigarette smoke roll over the top of the glass into your area... I only went once because of that.
I suspect they made this public because many customers will notice that they are no longer carrying Adafruit products. I respect both companies greatly and have purchased from them in the past. It will be interesting to see what happened, if that is made public.
This is really cool! As someone who has basically no piano training, this is fun! Perhaps there could be some super-easy mode where you actually highlight the keys while you're playing the sounds (in simon mode) to help the super noobs train their ears.
I think there is also some FOMO involved. Once people started saying how AI was helping them be more productive, a lot of folks felt that if they didn't do the same, they were lagging behind.
Do they actually have access to that info "in-band"? I would guess not. OTOH it should be straightforward for the LLM program to report this -- someone else commented that you can do this when running your own LLM locally, but I guess commercial providers have incentives not to make this info available.
Naturally, their "confidence" is represented as activations in layers close to output, so they might be able to use it. Research ([0], [1], [2], [3]) shows that results of prompting LLMs to express their confidence correlate with their accuracy. The models tend to be overconfident, but in my anecdotal experience the latest models are passably good at judging their own confidence.
I don't believe that it just analyzes the transcription. I asked Gemini to look at the youtube video referenced on the site below and "build" something that duplicates that device. It did a pretty good approximation that it could not have done without going through the full video.
The author actually discusses the results of the paper. He's not some rando but a Wharton Professor and when he is comparing the results to a grad student, it is with some authority.
"So is this a PhD-level intelligence? In some ways, yes, if you define a PhD level intelligence as doing the work of a competent grad student at a research university. But it also had some of the weaknesses of a grad student. The idea was good, as were many elements of the execution, but there were also problems..."
100%. I have known a couple of people that did some form of "medical tourism", mostly for expensive dental work. In both cases they did some form of tech contract work as a sole proprietorship and bought their own health insurance (not through a partner). The overlap of people who can save up thousands of dollars for treatment abroad and have poor health insurance is probably not too large.
reply