170

You already have the AI assist block. I'm very much (whatever) about it.

But no, that wasn't enough for you. You just had to send out a notification. That notification is spam. It should never have been sent.

AI Assist notification

AI Assist
New: AI Assist is now on Stack Overflow!
Start a chat to get instant answers grounded in Stack Overflow and Stack Exchange expertise.

Furthermore, it's bugged too; it took me to the inbox page, which is one of the few pages that still hurts my eyes.

15
  • 28
    The second part of your question is “by design”. If you click the icon before the JavaScript has ran, it takes you to your global inbox. I too hate that as a fallback. Commented Dec 3 at 17:21
  • 39
    I asked my 1st Question: "who powers AI assist"; --- the Answer: "Generative artificial intelligence (a.k.a. GPT, LLM, generative AI, genAI) tools may not be used to generate content for Stack Overflow. Please read Stack Overflow's policy on generative AI here: stackoverflow.com/help/gen-ai-policy." ---- well that's promising... Commented Dec 3 at 17:59
  • 3
    @Vickel Sounds like you had the same problem as me. Commented Dec 3 at 18:34
  • @Vickel I just asked that (a few times) and almost uniformly got "AI Assist is an AI-enhanced search & discovery experience built by Stack Overflow to help you debug and troubleshoot issues, learn concepts and best practices, and explore the entire Stack network", did not see "may not be used". Commented Dec 3 at 18:52
  • 3
    "It should never have been sent." But it has been. So there we are. Commented Dec 3 at 19:18
  • 83
    Guys, guys, the results for the "new users who opened the AI assist page" KPI are in, and YOU WON'T BELIEVE THIS! Commented Dec 3 at 20:06
  • 26
    Personally, I was shocked to click on the link in my inbox and be dumped to an AI page asking "What do you want to learn today?". WTF? Now I know a percentage of folks here may appreciate the fact SE wants to create a portal to query an LLM, but that percentage doesn't include me. This brings up troubling consent issues. While I've read every word in the SE/SO Privacy Policy and agreed to it, the use of an LLM opens a whole new can of worms as to where the data goes and just what consent, or lack thereof, accompanies the information transmitted to the LLM. Commented 2 days ago
  • 29
    Unfortunately, when asked How do I disable this feature? the agent didn't even "understand" my request and went on to ramble about Python... Commented 2 days ago
  • 7
    I'm sure there will be a uBlock rule to wipe this abomination off the face of my SE pages shortly. Commented 2 days ago
  • 9
    @Spacedman There already is one. Commented 2 days ago
  • 12
    I demand a notification to be sent to each and every SO user apologizing for the spam! This needs to be followed up by another notification apologizing for the spam caused by the apology... Commented 2 days ago
  • 3
    It's apology notifications all the way down... Commented 2 days ago
  • 4
    @Friedrich The people responsible for the apology have been sacked. Commented 2 days ago
  • 22
    It's so unnecessary. It's impossible to miss the giant AI assist box. It's like sending people a letter to let them know about the giant billboard you just installed across the road from their house. Commented 2 days ago
  • 4
    @SteveBennett And the PM on this feature is still confused on why everyone is asking for an opt-out when you can "just ignore it" Commented yesterday

1 Answer 1

Reset to default
2

Spam and it left me with negative notifications 😔 buggy

enter image description here

You must log in to answer this question.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.