4 ms·

I don't see "AI sometimes gets thing absurdly wrong" getting fixed any time soon.

dainiusse 2y ago
Llm's to be fair, not AI
stingraycharles 2y ago
Google’s spam filter is pretty good though and has been for a long, long time. But I guess we need to sprinkle everything with a bit of extra LLM AI these days.
- mrjin 2y ago
  Did not work for me from the very first day. I was one of Gmail Beta testers, unfortunately, my new account started receiving one the very first day I registered it. I asked for a blacklist and kept being pushed back saying their filter was good enough, and I should never have never needed a blacklist. Oh well.
  - helsinkiandrew 2y ago
    > Did not work for me from the very first day. I was one of Gmail Beta testers But now its superb - reporting that a mail is spam does a good job of marking future mails from that sender as spam and moving messages from spam folder to inbox does the opposite.
    - dagmx 2y ago
      I wish I could say the same. I mark so many emails in my Gmail as phishing attempts but it just never learns. They’re super obvious ones too with a nonsensical email address, a repeating pattern about mcaffee or Norton in the title and an almost empty body with a pdf attached. Meanwhile Gmail also happily never learns when I tell it something isn’t spam either.
      - mrjin 2y ago
        That has been there since the very first day. It seems nothing has changed since.
    - mrjin 2y ago
      Doesn't matter, I no longer want to use google products.
- beardyw 2y ago
  Agreed. I haven't seen spam in a long time.
  - joking 2y ago
    I have to see spam not because it passed the filter, but I have to check the spam folder weekly as some legitim emails end there
- jeroenhd 2y ago
  Google's spam filter lets through tons of spam for me. I think some spammers are abusing Google's weird DKIM configuration to send me emails that were supposedly sent from my own email address. No amount of clicking "report spam" will do anything. It also blocks just about any small domain that emails me for the first time. No amount of SPF or DKIM will convince Google that you're legitimate party, there's some kind of minimal volume you need to send Google to make your emails arrive to Gmail inboxes the first time. It works when it works, but when it doesn't, it's broken without repair. It works _most of the time_ and it's better than Outlook (though that's not a high bar to clear).
- thaumasiotes 2y ago
  > Google’s spam filter is pretty good though and has been for a long, long time. What? This hasn't been true for at least 15 years. Instead, Google's spam filter is far, far more aggressive than could conceivably be appropriate, and it routinely filters important communications from people you know.
- acdha 2y ago
  I used Gmail from the first public signups to their bitchedw Apps for Domains migration. After we switched to Fastmail, the first thing we noticed was how much less spam we were seeing and the second was how many legitimate messages had been incorrectly filtered by their priority inbox system.
xanderlewis 2y ago
‘Absurdly wrong’ is spot on. The issue with AI isn’t that it simply gets things wrong — as is frequently pointed out, so do humans. The issue is that it gets things wrong in a way that comes out of nowhere and doesn’t even have a post-rationalised explanation. The big claim about AI systems (especially LLMs) is that they can generalise, but in reality the ‘zone of possible generalisation’ is quite small. They overfit their training data and when presented with input out of distribution they choke. The only reason anyone is amazed by the power of LLMs is because the training set is unimaginably huge. In fifty years we’ll have systems that make this stuff look as much like ‘AI’ as, say, Djikstra’s algorithm does now.
- bitwize 2y ago
  I first noticed this with Watson (IBM's language processing system) when it played Jeopardy!: when it was right, it was spot on and usually faster than the human contestants; but when it was wrong, it was way, way off base. Part of that has to do with the fact that language is not the same for an LLM as it is for a person. If I say to you the sentence "The cat sat on the mat", that will evoke a picture, at the very least an abstract sketch, in your mind based on prior experience of cats, mats, and the sitting thereupon. Even aphantasic people will be able to map utterances to aspects of their experience in ways that allow them to judge whether something makes sense. A phrase like "colorless green dreams sleep furiously" is arrant nonsense to just about everybody. But LLMs have no experiences. Utterances are tokens with statistical information about how they relate to one another. Nodes in a graph with weighted edges or something. If you say to an LLM "Explain to me how colorless green dreams can sleep furiously", it might respond with "Certainly! Dreams come in a variety of colors, including green and colorless..." I've always found Searle's argument in the Chinese Room thought experiment fascinating, if wrong; my traditional response to it was "the man in the room does not understand Chinese, but the algorithm he's running might". I've been revisiting this thought experiment recently, and think Searle may have been less wrong than I'd first guessed. At a minimum, we can say that we do not yet have an algorithm that can understand Chinese (or English) the way we understand Chinese (or English).
BeFlatXIII 2y ago
Breaking news: the AI made another absurd mistake

Llm's to be fair, not AI

Google’s spam filter is pretty good though and has been for a long, long time. But I guess we need to sprinkle everything with a bit of extra LLM AI these days.

mrjin 2y ago
Did not work for me from the very first day. I was one of Gmail Beta testers, unfortunately, my new account started receiving one the very first day I registered it. I asked for a blacklist and kept being pushed back saying their filter was good enough, and I should never have never needed a blacklist. Oh well.
- helsinkiandrew 2y ago
  > Did not work for me from the very first day. I was one of Gmail Beta testers But now its superb - reporting that a mail is spam does a good job of marking future mails from that sender as spam and moving messages from spam folder to inbox does the opposite.
  - dagmx 2y ago
    I wish I could say the same. I mark so many emails in my Gmail as phishing attempts but it just never learns. They’re super obvious ones too with a nonsensical email address, a repeating pattern about mcaffee or Norton in the title and an almost empty body with a pdf attached. Meanwhile Gmail also happily never learns when I tell it something isn’t spam either.
    - mrjin 2y ago
      That has been there since the very first day. It seems nothing has changed since.
  - mrjin 2y ago
    Doesn't matter, I no longer want to use google products.
beardyw 2y ago
Agreed. I haven't seen spam in a long time.
- joking 2y ago
  I have to see spam not because it passed the filter, but I have to check the spam folder weekly as some legitim emails end there
jeroenhd 2y ago
Google's spam filter lets through tons of spam for me. I think some spammers are abusing Google's weird DKIM configuration to send me emails that were supposedly sent from my own email address. No amount of clicking "report spam" will do anything. It also blocks just about any small domain that emails me for the first time. No amount of SPF or DKIM will convince Google that you're legitimate party, there's some kind of minimal volume you need to send Google to make your emails arrive to Gmail inboxes the first time. It works when it works, but when it doesn't, it's broken without repair. It works _most of the time_ and it's better than Outlook (though that's not a high bar to clear).
thaumasiotes 2y ago
> Google’s spam filter is pretty good though and has been for a long, long time. What? This hasn't been true for at least 15 years. Instead, Google's spam filter is far, far more aggressive than could conceivably be appropriate, and it routinely filters important communications from people you know.
acdha 2y ago
I used Gmail from the first public signups to their bitchedw Apps for Domains migration. After we switched to Fastmail, the first thing we noticed was how much less spam we were seeing and the second was how many legitimate messages had been incorrectly filtered by their priority inbox system.

xanderlewis 2y ago

‘Absurdly wrong’ is spot on. The issue with AI isn’t that it simply gets things wrong — as is frequently pointed out, so do humans. The issue is that it gets things wrong in a way that comes out of nowhere and doesn’t even have a post-rationalised explanation. The big claim about AI systems (especially LLMs) is that they can generalise, but in reality the ‘zone of possible generalisation’ is quite small. They overfit their training data and when presented with input out of distribution they choke. The only reason anyone is amazed by the power of LLMs is because the training set is unimaginably huge. In fifty years we’ll have systems that make this stuff look as much like ‘AI’ as, say, Djikstra’s algorithm does now.

bitwize 2y ago
I first noticed this with Watson (IBM's language processing system) when it played Jeopardy!: when it was right, it was spot on and usually faster than the human contestants; but when it was wrong, it was way, way off base. Part of that has to do with the fact that language is not the same for an LLM as it is for a person. If I say to you the sentence "The cat sat on the mat", that will evoke a picture, at the very least an abstract sketch, in your mind based on prior experience of cats, mats, and the sitting thereupon. Even aphantasic people will be able to map utterances to aspects of their experience in ways that allow them to judge whether something makes sense. A phrase like "colorless green dreams sleep furiously" is arrant nonsense to just about everybody. But LLMs have no experiences. Utterances are tokens with statistical information about how they relate to one another. Nodes in a graph with weighted edges or something. If you say to an LLM "Explain to me how colorless green dreams can sleep furiously", it might respond with "Certainly! Dreams come in a variety of colors, including green and colorless..." I've always found Searle's argument in the Chinese Room thought experiment fascinating, if wrong; my traditional response to it was "the man in the room does not understand Chinese, but the algorithm he's running might". I've been revisiting this thought experiment recently, and think Searle may have been less wrong than I'd first guessed. At a minimum, we can say that we do not yet have an algorithm that can understand Chinese (or English) the way we understand Chinese (or English).

BeFlatXIII 2y ago

Breaking news: the AI made another absurd mistake