Live data from Hacker News

Google Workspace thinks my domain is an email provider (2025)

blog.elis.cc

121–130 of 163 posts

Re: Google Workspace thinks my domain is an email provider (2025)

#121

> 3.) Fictitious Domain Considerations: As you may know, domains that has ".web" have faced issues related to fictitious domain name usage. While ".one" is not ".web" the combination of the subdomain and TLD triggers similar security protocols. This is an LLM response. It's the classic pattern where the LLM says something completely unrelated to the topic at hand, realizes that it doesn't have a backspace key, and tr…

Has anyone ever tried giving LLMs a backspace? I have long wondered if their responses could be better if they had the ability to correct earlier parts of their text. Imagine how much harder it would be for us to write coherently with no backspace!

ChatGPT does this. It will present an initial response, then change its mind partway through the thinking process and present a different response.

Re: Google Workspace thinks my domain is an email provider (2025)

#122

Earlier quoted context omitted.

Has anyone ever tried giving LLMs a backspace? I have long wondered if their responses could be better if they had the ability to correct earlier parts of their text. Imagine how much harder it would be for us to write coherently with no backspace!

Thinking mode is exactly this. The LLM is allowed to dump thoughts, correcting itself along the way, and only surfaces the cleaned-up final answer to the user

It's not exactly this. It's something like it. But it's not a token that erases a previous token. Not that I know how that would work or how you'd get training data (edit histories of internet comments seem too few).

(You could have a token that hides previous tokens, but that'd be rather closer to CoT.)

Re: Google Workspace thinks my domain is an email provider (2025)

#123

Earlier quoted context omitted.

Thinking mode is exactly this. The LLM is allowed to dump thoughts, correcting itself along the way, and only surfaces the cleaned-up final answer to the user

It's not exactly this. It's something like it. But it's not a token that erases a previous token. Not that I know how that would work or how you'd get training data (edit histories of internet comments seem too few). (You could have a token that hides previous tokens, but that'd be rather closer to CoT.)

Backtracking, of sorts. The LLM descends a seemingly fruitful path, and all of a sudden that path seems less so. So you backtrack a few tokens ("erasing"/"undoing") and resample the probability distribution a bit back in the stream.

Re: Google Workspace thinks my domain is an email provider (2025)

#124

Earlier quoted context omitted.

Thinking mode is exactly this. The LLM is allowed to dump thoughts, correcting itself along the way, and only surfaces the cleaned-up final answer to the user

It's not exactly this. It's something like it. But it's not a token that erases a previous token. Not that I know how that would work or how you'd get training data (edit histories of internet comments seem too few). (You could have a token that hides previous tokens, but that'd be rather closer to CoT.)

You could burn a few tokens by prompting the LLM to summarize the omit all of the "wrong" answers (where it overrode itself), after the user accepts the response. It'd probably reduce token use in the end, especially for long sessions.

Re: Google Workspace thinks my domain is an email provider (2025)

#125

> 3.) Fictitious Domain Considerations: As you may know, domains that has ".web" have faced issues related to fictitious domain name usage. While ".one" is not ".web" the combination of the subdomain and TLD triggers similar security protocols. This is an LLM response. It's the classic pattern where the LLM says something completely unrelated to the topic at hand, realizes that it doesn't have a backspace key, and tr…

To do this you need to run

    hallucinated command
but wait, this command only existe for aws cli, not for oracle cli, so you need to run

    actual command
Regularly happens to me in chatgpt.

Re: Google Workspace thinks my domain is an email provider (2025)

#126

Earlier quoted context omitted.

Has anyone ever tried giving LLMs a backspace? I have long wondered if their responses could be better if they had the ability to correct earlier parts of their text. Imagine how much harder it would be for us to write coherently with no backspace!

ChatGPT does this. It will present an initial response, then change its mind partway through the thinking process and present a different response.

Oh wow, I'd like to see this. Any prompt which reliably triggers it? (Seahorse emoji comes to mind!)

Re: Google Workspace thinks my domain is an email provider (2025)

#127

Earlier quoted context omitted.

Has anyone ever tried giving LLMs a backspace? I have long wondered if their responses could be better if they had the ability to correct earlier parts of their text. Imagine how much harder it would be for us to write coherently with no backspace!

ChatGPT does this. It will present an initial response, then change its mind partway through the thinking process and present a different response.

Copilot does this too, kinda. It will give an initial response, realize that it might tangentially criticize Microsoft, then change that response to a refusal.

Re: Google Workspace thinks my domain is an email provider (2025)

#128

The end is really hilarious. Google had a stupid frontend-only validation, it seems.

Since Google themselves claim that's a security check, there's a bug bounty here. The author of that article missed their chance making Google eat their words.

Or he missed the chance to get Google to move that check to the back end and then when it is time to renew find out that he cannot, if he’s lucky enough for it to not break earlier.

Re: Google Workspace thinks my domain is an email provider (2025)

#129

Earlier quoted context omitted.

Thinking mode is exactly this. The LLM is allowed to dump thoughts, correcting itself along the way, and only surfaces the cleaned-up final answer to the user

It's not exactly this. It's something like it. But it's not a token that erases a previous token. Not that I know how that would work or how you'd get training data (edit histories of internet comments seem too few). (You could have a token that hides previous tokens, but that'd be rather closer to CoT.)

I'm not sure how that would work. You'd need to be able to reverse the operations in the KV cache, and I'm not sure if it's doable or not. And then you'd need to pick another path, or "rebalance" the probabilities or you'd go the same token path. CoT is probably easier to implement (but not necessarily better).

Re: Google Workspace thinks my domain is an email provider (2025)

#130
With how many products Google suddenly retires, and how they often just ban or remove accounts without any sort of support, it’s mind boggling to me that anybody still considers to use any Google service honestly.

Well, apart from search and advertisement, maybe.

Post reply on HN