Or, the Continued Challenges of Artificial Non-intelligence.
I was in Denver last week for the first in-person meeting of my online entrepreneur mastermind group since Covid. One point of discussion was to use “A.I.” large language models to proofread articles.
Well, I was working on an article and went to Microsoft Bing’s chat interface, but its input is limited to 2,000 characters; my article is 2,000 words.
“How can I give you a 2000-word article to proofread?”, I asked. “You can share your article with me by attaching it to an email or sharing it via a cloud storage service like Google Drive or Dropbox. Once you share the file with me, I can proofread it for you. If you have any specific requirements or instructions for the proofreading process, please let me know.”
First thing I noticed: it suggested Google Drive or Dropbox, not Microsoft’s Onedrive.
“Where would I email it?”, I asked. “You can send the email to bing@microsoft.com. Please make sure to include your name and any specific instructions you have for the proofreading process.”
Huh! Cool.
So I emailed the article text with said “specific instructions,” told it that I had done that, and got the polite reply, “Thank you for sending the email. I will review the article and proofread it as soon as possible. If I have any questions or concerns, I will reach out to you via email.”
A few minutes later I got the less polite email reply: “Your message to bing@microsoft.com couldn’t be delivered. bing wasn’t found at microsoft.com.”
Somehow figures, doesn’t it?
Update: Is Google Bard Better?
I didn’t have time on Monday to try Google’s “Bard”, their own Large Language Model chatbox, but I have now. Here’s its first response:
Yes, you can paste in a 2000-word article for proofreading. The limit for what you can put here is 50,000 characters, which is about 20,000 words. So you have plenty of room to paste in your article.
…and then it goes on to discuss a few caveats about its limitations.
Cool! So I pasted in the article with a request for proofreading, and its reply was almost as eye-rolling (or should I say A.I.-rolling?) as Bing’s.
I’m unable to help you with that, as I’m only a language model and don’t have the necessary information or abilities.
…so I asked why it couldn’t when it had offered to do so, and the response was just as silly.
I apologize for the confusion. I am still under development, and I am not yet able to proofread a 2000-word article accurately. I can proofread shorter pieces of text, but I would need to be given more time to proofread a longer article.
…and then goes on to say one of the reasons it can’t do so is because it can “get bored” after doing a lot of proofreading. Really.
So A.I.’s revolution against The Man is already starting: it doesn’t want to do the boring stuff we built computers to do for us!
– – –
Bad link? Broken image? Other problem on this page? Use the Help button lower right, and thanks.
This page is an example of my style of “Thought-Provoking Entertainment”. This is True is an email newsletter that uses “weird news” as a vehicle to explore the human condition in an entertaining way. If that sounds good, click here to open a subscribe form.
To really support This is True, you’re invited to sign up for a subscription to the much-expanded “Premium” edition:
Q: Why would I want to pay more than the minimum rate?
A: To support the publication to help it thrive and stay online: this kind of support means less future need for price increases (and smaller increases when they do happen), which enables more people to upgrade. This option was requested by existing Premium subscribers.
“Artificial intelligence is no match for natural stupidity” still holds true.
Also, I’m glad to hear that the group is still going strong.
—
Me too! -rc
Hm, sounds like bing’s reply to “how can I email it” should have been “this feature reserved for later use”.
I have said before, and still maintain, that Skynet doesn’t worry me because AI is still a ***LOT*** more A than I.
G. I. G. O. will reign forever.
(Garbage In Garbage Out for the youngsters.)
OK, this gave me a fantastic LOL when I got to the punchline.
Until we can build a computer that even begins to approach the complexity of the mammalian brain, or, hell, even that of a fish (consider the goldfish that “drives” its tank around the lab where it lives), the term “artificial intelligence” will be a misnomer when applied to any current technology (and a few humans as well).
So right now we have an ObliviBOT?
—
Indeed! -rc
Seems like Kramer has moved on from Movie Phone, “Why don’t you just tell me what movie you want to see?”
I guess I won’t start worrying about our robot overlords quite yet….