Kinda like "I couldn't care less" vs "I could care less"?

Simple question here. If AI is writing code, then how is it executing this code the test it? If I have a smallish repo containing 250k lines of code, how is the AI executing what it has written that works within that codebase? Is it compiling an EXE, and running it within a Windows environment? Is it running PHP code on a LAMP stack? How does it test the query on my database containing 5 million records? How does it verify the HTML and CSS it created behaves correctly in half a dozen major browsers on desktop, tablet and phone?

We are talking about a Chinese State-owned company here, right? And you are really believing they are being open and honest? Exactly would a communist-owned company gain by being open and honest?

How do you know there haven't been thousands of other people killed due to this autopilot failing in China so far?

They are stating that the "public network" count includes private chargers, but they are shared by more than 1 person in some way. IE a business may have a charging station that can be used by some specific employees (and this is probably the greatest contributor to that "public" count). There might be an apartment building with 2 or 3 units, that has a charger that can only be used by the tenants at that apartment. Or potentially a homeowner has a charger that they share with their immediate neighbor. Those are not public chargers, but they are counted as such. They should be counted in the 700,000 private chargers instead, because they are by definition not public.

The public count should ONLY include a charger that any random person can drive up to and use. That is what has to be compared against "gas nozzles" which are entirely for public use.

I read TFA, and articles it references, but can't find the definition of "shared" private EV chargers nor a breakdown of how many of those 178k chargers are private. Considering "gas nozzles" are public, comparing them to some unknown number of "shared private" chargers, that potentially only a few people might be allowed to use each, is kind of misleading.

My son was interested in possibly getting a K truck, so we drove about an hour to a place in North Carolina that imports them. They have van-style, truck-style, and even fire trucks and versions with a beds that lift / dump. They are all right-hand drive. I think most were $8k, with fancier ones going for $10-$11k. They are made by a few different companies (Mitsubishi, Suzuki, Honda, Daihatsu), and each brand has its own little stylistic details.

Unfortunately my son is 6'4" and just cannot fit in these things. We thought about getting a van type, and moving the seat mounts back, but even then there just isn't enough room for his legs between the steering wheel and the door.

These things are getting popular in this region, especially for farm use, and they can be registered like any other vehicle for street use. In fact, the guy / business that's importing them has been lobbying hard and got some positive results about registering and using them in the US. One of the keys to this is they have to be at least a certain age (maybe 20 years or older?) He's got at least a couple hundred of them in stock. Apparently you can buy them REALLY cheap in Japan - a few thousand bucks. But, by the time you ship them, have an agent receive them, customs, and then some kind of inspection / paperwork for use in the USA, you're around the $7k mark total. So it's easier to shop them from guys like this who have also tested them and fixed up odds and ends. He keeps them all unlocked with keys in them, so you just hop in one and give it a go. Really low key and convenient way to check them out.

Anyway, they are importable and somewhat popular in this part of the USA.

Oh it will be the insurance companies pushing for this. They already have devices you install that monitor your driving to get the better rates.

Note that you wouldn't get lower rates from scanning your license - you just won't be penalized with higher rates. If that makes sense.

In my state (Virginia) state law allows insurance agencies to automatically put anyone at your address on your policy automatically. My wife and I have separate policies (we both have kids but none together). Our insurance companies (Erie and Progressive) try to screw us over regularly - we both have had each other's driving age kids automatically added to our policies, which of course can greatly increase the cost, and then have to jump through hoops sending them documentation proving they are actually covered on another policy. Then they are converted to a "List Only" driver, so they are on the policy, but not considered a driver.

So you'd better believe insurance companies would love to track which specific person is driving which vehicle and how much, and bill your accordingly. "We see that this 17 year old male has driven your vehicle more than once in a month. We've added them to your policy as a driver of that vehicle. You own $1,250 for the remainder of your policy term."

There's another factor which is skill. There is art (paintings, drawings, sculptures, music, you name it) that has been created that only an extremely, extremely tiny fraction of humanity could possibly reproduce, no matter how much time or effort they exerted. To AI it's just pixels, so it could easily mimic things humans have created in a way that few other humans can.

A token is a portion of a word. It doesn't equate to individual letters, nor entire words, but something in between. The average is around .75 tokens per word. So four "average" words take 5 tokens.

So 20 tokens per second is perfectly fine for a single person interactively chatting with the LLM. If you're doing any sort of larger data processing (feeding in large documents, outputting large documents, or multiple users) it's pretty slow.

Yes, especially since Google removed one of the most useful search features it ever had, which was the ability to view the cached page from the last time Google crawled it.

2.9 billion people may have had Social Security numbers, other financial data compromised It was actually around 900 million SSN numbers leaked.

Huge data breach involving social security numbers could impact millions of Americans This separate leak had 270 million SSNs.

A leak with hacked Social Security numbers is stoking concern. "Security analysts who spoke with NBC News also said that while it is still concerning, the leak of Social Security numbers isn’t a reason for panic — many people’s numbers have already been leaked in previous hacks."

Isn't he relinquishing his federal power over education by dismantling the federal department of education? I mean if that's his goal, to rewrite history and control what is taught, wouldn't he want to consolidate power at his level of the government even more?

I just don't understand the logic here, because he wouldn't need to return control to the states. It is totally within his power to have the education department let the states do whatever they want (and for the most part they already can, right?) and continue spending tons of federal money on the department.

How does dismantling the federal DOE make states like California teach what Trump wants them to? I'm legitimately curious how dismantling the DOE will impact what my kids are taught in school. If there's something I'm missing here I'd like to know what it is, because this power could be wielded by either side down the road when we have a different president.

He's been saying he wants to do this all along. When he was president in 2018 he wanted to release them, but after review they decided to wait 3 more years (which was beyond his first term) to revisit the decision. Most likely because people were still living that could be impacted by information they wanted to release.

He said it multiple times during his most recent campaign.

Then the order to release them was signed almost two months ago.

This isn't something totally out of the blue, but a topic he's been talking about for almost a decade now. He's a lame duck president, and clearly Trump doesn't care whether or not the "masses are getting grumpy". He's obviously doing whatever he wants to do.

So you have AI crank out images of unique things... humans, vehicles, mascots, building architecture, every permutation of music you can come up with, and copyright it all. Then go after anything similar claiming it is a derivative work. This is just one reason why we can't have AI generated stuff become copyrighted.

AI is a tantalizing tease. You can see glimpses of what it can potentially do, but then it hallucinates and produces something beyond useless and potentially dangerous.

As a software developer using AI, it's clear to me when AI produces garbage or something not viable, and I'll ignore what it suggested or if I'm prompting I'll steer it in the right direction. But I'm using 35 years of experience to filter out the garbage it produces. It can speed up what I do, generally by saving me some time digging through API documentation and the like, so I use it, but I do so with caution.

Now when you throw this tech to the masses and expect it to work correctly 100% of the time, that's another story. If I tell Alexa or Siri to turn something on or turn it off or set a timer, vacuum my floor, close the garage door, there is no room for interpretation there. It either understood my speech and needs to do exactly what I said, or else ask for clarification or just flat-out do nothing and ignore what I said. The danger with AI is it making an assumption or hallucinating something, and opening my garage door while I'm on vacation because I said my car is too hot and Alexa assumed it was still in my garage and thought maybe opening the garage door would cool it off.

Back in the late 1980s, I was messing around with a tantalizing little piece of tech, which was an IC chip Radio Shack sold for about $15 that could recognize speech. The possibilities of this thing were pretty cool, and so I began experimenting with it. After a bit I realized it was pretty much a worthless gimmick - because almost everything you said was converted ("best fit") into one of the handful of words it was trained to recognize. I don't remember the exact words, but they were things like "yes", "no", "on", "off", "go", "stop", etc. Since the algorithm would almost always convert what you said into one of the words it recognized, it really only worked in a totally quiet environment where the only words you said were the ones the device understood, otherwise it would constantly be triggering from words that didn't even sound close to what it thought they were.

AI has been like this for a long time, and has "tricked" a lot of researchers over the last half century. At first it was always hardware limitations - if only we had more ram, more training data, faster processing - but now it's a fundamental misunderstanding by a lot of people exactly what LLMs are doing, and what they really should be used for.