𝙲𝚑𝚊𝚒𝚛𝚖𝚊𝚗 𝙼𝚎𝚘𝚠

The PR had some issues regarding files that were pushed that shouldn’t have been, adding refactors that should have been in separate PRs, etc…

Though the main reason is that Signal doesn’t consider this issue a part of their threat model.

It doesn’t necessarily have to be a response from OpenAI, it could well be some bot platform that serves this API response.

I’m pretty sure someone somewhere has created a product that allows you to generate bot responses from a variety of LLM sources. And if whatever is interacting with it is simply reading the response body and stripping out what it expects to be there to leave only the message, I could easily see a fairly bad programmer create something that outputs something like this.

It’s certainly possible this is just a troll account, but it could also just be shit software.

It can’t be effective. The risk of false-positives is huge.

Perhaps it was being influenced by the chat history. But try asking how many r’s in raspberry, it does get that consistently wrong for me. And you can ask it those followup questions to easily get it to spout nonsense, and that was mostly my point; figuring out if you’re talking to an LLM is fairly trivial.

My point is that telling it a right answer is wrong often causes LLMs to completely shit the bed. They used to argue with you nonsensically, now they give you a different answer (often also wrong).

The only question missing at the start was "How many r’s are there in the word ‘veryberry’. I think raspberry also worked when I tried it. This was ChatGPT4-O. I did mark all the answers as bad, so perhaps they’ve fixed this one by now.

Still, it’s remarkably trivial to get an LLM to provide a clearly non-human response.

Here’s what I got:**

It’s dead simple to see if you’re talking to an LLM. The latest models don’t pass the Turing test, not even close. Asking them simple shit causes them to crap themselves really quickly.

Ask ChatGPT how many r’s there are in “veryberry”. When it gets it wrong, tell it you’re disappointed and expect a correct answer. If you do that repeatedly, you can get it to claim there’s more r’s in the word than it has letters.

The top two candidates may be quite polarising for example, whereas a 3rd candidate might be acceptable to all. Or there could be a large group of quite mediocre candidates.

It’s not bad way per sé, but it’s basically a band-aid around FPTP, instead of fixing the problem properly.

France has a slightly odd election system, where the vote for president is a free-for-all initially, and then the top two candidates face off.

He’s currently the second largest party, so he’s banking on being the contender against Le Pen for president, hoping that the left, center and moderate right help him win again (because Le Pen would be worse than he’d be).

It’s not a terrible strategy, but it is a little risky.

Any enterprise working with sensitive data certainly has to disable the feature. And turns out, that’s most enterprises.

I have heard very little, if any, enthusiasm about this. Nobody seems to be excited about it at all.

Nuclear reactors are ill-suited for baseloads, because they can’t scale their output in an economical way.

You always want the cheapest power available to fulfill demand, which is solar and wind. Those regularly provide more than 100% of the demand. At this point, any other power sources would shut off due to economical reasons. Same with nuclear, nobody wants to buy expensive nuclear energy at peak solar/wind hours, so the reactor needs to turn off. And while some designs can fairly quickly power down, powering up is a different matter and doing either in an economically feasible way is a fantasy right now.

If solar and wind don’t provide enough power to satisfy demand, some other power source needs to turn on. Studies have already shown that current-gen battery storage is capable of doing so. Alternatives could be hydrogen or gas power stations. Nuclear isn’t an option economically speaking.

It sounds more like Rashida Jones.

The polls were accurate, but right near the end Yesilgöz made the critical mistake of publicly stating she’d be willing to work with the PVV. This catapulted the PVV to the top, as he was suddenly a viable alternative to VVD/NSC/BBB for a lot of voters.

We saw this sudden rise in the polls as well, with the PVV suddenly in first right before the election. And that prospect, the PVV being the largest, drew in some more voters to them.

Polls are projections, but they can’t factor in unforeseen events between the poll and the election.

Their win is not that huge as far as current projections go. They’ll grow, but the likely coalition is the current one.

Ah, that may be true indeed.

Signal recently updated to allow usernames instead of phone numbers.

A strong bias against genocide and in favour of peace is not a very disagreeable one.

Techwise it probably doesn’t, but then there’s marketeers, sales, accountants, legal, etc…

Well, more RAM will always help. An iPhone with more RAM will allow it to perform better than one with less RAM. Similarly, too little RAM will hamper performance regardless of the device.

Android has a garbage collector, meaning it requires an additional 2GB of RAM of overhead to keep things smooth. iPhones run significantly hotter than Androids, and consume more energy to achieve their performance gains.

It’s not true to simply state “one is better than the other”. There’s various metrics in which either one may be better.