See if Tesla does well that proves musk is a genius and should be paid more. If it does badly it proves he was unmotivated and should be paid more.

Yeah as someone that lives in a city with mass transit, you change your habits.

You shop two or three times week at somewhere in walking distance. You walk to the vet, and you order lumber online with next day delivery.

If I genuinely need a car, there’s one parked in the next street I can rent with an app.

On top of that parking here is a pain in the arse, and the average traffic speed is something like 7mph.

Well it is trained to copy musk.

I guess it might work if HR don’t know how an LLM works. There’s not many that can edit a word file so it includes whited-out footnotes.

You’re better off getting a friend to lie for you. They can say they added it while helping you with formatting and you know nothing about it.

Genuinely, this already happens in large companies for related reasons.

The CV is on file, and if HR reprocess it for any reason e.g. relocation or change of role, it’s automatic dismissal for dishonestly if they catch a deliberate lie.

No but they can fire you later even if you’re good at the job.

Then you’re stuck in an even worse position with a big gap in your CV and no reference.

Unfortunately, this is seen as dishonestly and is grounds for immediate dismissal in a lot of places.

Right. That’s why they overreact to everything, and bring old military equipment on swat raids.

They’re much more likely to panic and drive an APC through the crowd or return fire on a mostly unarmed crowd using automatic weapons.

Just ask yourself, “what has Israel done recently?” and remember that US police train with them.

I live in a walkable European city.

My nearest library is 5 minutes away, there’s a bigger library maybe 20 minutes away, and for anything further I’d take public transport.

Worse, it seems to be Tony Blair.

Id cards were one of Blair’s most unpopular policies, and all the AI first stuff seems to be coming straight from the Tony Blair institute.

In practice it’s very systematic for small networks. You perform a search over a range of values until you find what works. We know the optimisation gets harder the deeper a network is so you probably won’t go over 3 hidden layers on tabular data (although if you really care about performance on tabular data you would use something that wasn’t a neural network).

But yes, fundamentally, it’s arbitrary. For each dataset a different architecture might work better, and no one has a good strategy for picking it.

Probably because there’s no good reason.

At least one intermediate layer is needed to make it expressive enough to fit any data, but if you make it wide enough (increasing the blobs) you don’t need more layers.

At that point you then start tuning it /adjusting the number of layers and how wide they are until it works well on data it’s not seen before.

At the end, you’re just like “huh I guess two hidden layers with a width of 6 was enough.”

That’s funny, because there’s also emoji associated with Trump.

🤏🍊🍆

You can look up what Adobe are doing in this space.

Long-term it’s going to be something like every secure device comes with its own inbuilt unit for cryptographically signing the raw files. These signatures can then be matched against manufacturers databases of approved signatures.

This doesn’t guarantee that nothing has been tampered with, but it does provide a link back to the original device allowing you to inspect it.

There are huge privacy concerns as well, anything that’s used to indicate authenticity can be used to track.

It’s interesting. There’s a lot of talk about how chatgpt makes people lazy, but honestly I think Google killed the “read the manual” ethos.

Back in the day when you couldn’t just search for everything, you needed enough understanding of the manual to find anything in the index.

So a key part of figuring anything out was reading at least the start of the manual.

Now, fuck it, you just type into Google and try to guess enough context to understand what’s going on.

Yeah, but it can be filled in with “house”.

Hexbear is already flooded with beanis posts.

Looking forward to seeing beanis everywhere in the next version of Facebook’s LLM.

Garmin sends all your health data to the cloud and the app won’t work without an Internet connection.

On the plus side, they’re not part of the Google/Apple/Samsung data ecosystems, and I don’t think actually they do anything with the data, beyond computing statistics for you.

Depends how much you’re prepared to trust them I guess.

For Ubuntu you just run sudo apt install kde-plasma-desktop. Presumably it’s the same for the derivative you’re using.

After that gnome will still be there, and you just toggle between the two at login.

You’ve got it backwards.

This kind of voting forces the existence of two party systems.

Suppose you have two parties one left wing that gets 60% of the vote and one right wing that gets 40% in every district. Right now the left wing always wins.

If the left wing party splits into two blocks of 30% each, the right wing always wins.

So if you want to win, you can never split from the large parties, even if you feel unrepresented.