I would be wary of trying to use Google autocomplete for making any important decisions. Until they blacklisted showing autocomplete results for queries that look like "Why are <ethnicity>..." or "<ethnicity> are ...", you could get some interesting but certainly not trustworthy results.

They said it was informative, they didn't say anything about making important decisions.

It shows you what is often searched for which is informative, even if in that case it shows that racism is common (which arguably is informative within itself, albeit depressing).

Personally I like doing "[Something] Vs " to see what people consider are competitors. For example: "hacker news vs " returns:

- "hacker news vs reddit"

- "hacker news vs slashdot"

- "hacker news vs code"

etc.

I know you're mostly joking, but you'll get the same for any conspiracy:

Fluoride is ...

Fluoride is bad

Fluoride is toxic

Fluoride is poison

All of these are true! Except maybe the first one, which is a gross oversimplification.

"The dose makes the poison" is not something I expect an average Google user to have modeled accurately.

Edit: after reviewing other examples in this thread, my opinion of an average Google user has fallen drastically. Was mostly having a laugh at the fact fluoride salts are generally a thing I avoid, which seems to be the intuition represented by your example.

How did you manage to use emojis on HN?

I think HN stores text encoded as UTF-8, so all you need to do is input an emoji via your keyboard/copy-paste.