Sounds like that AI software needs yet another `if` clause...
...
except: print("I'm sorry, as an LLM")81–90 of 164 posts
Sounds like that AI software needs yet another `if` clause...
...
except: print("I'm sorry, as an LLM")In WWII in the Netherlands, when encountering a stranger, they'd have them pronounce 'Scheveningen' as a check-phrase to distinguish if they were dealing with a Dutch or German person. Now, we can ask random strangers on the internet to spell out some glitch tokens to determine if you're dealing with a LLM bot.
In WWII in the Netherlands, when encountering a stranger, they'd have them pronounce 'Scheveningen' as a check-phrase to distinguish if they were dealing with a Dutch or German person. Now, we can ask random strangers on the internet to spell out some glitch tokens to determine if you're dealing with a LLM bot.
> The Gileadites captured the fords of the Jordan leading to Ephraim, and whenever a survivor of Ephraim said, “Let me cross over,” the men of Gilead asked him, “Are you an Ephraimite?” If he replied, “No,” 6 they said, “All right, say ‘Shibboleth.’” If he said, “Sibboleth,” because he could not pronounce the word correctly, they seized him and killed him at the fords of the Jordan.
- Judges 12:5
In WW II, a well-known challenge/password/countersign set used by American and British soldiers during the D-Day landings in France was "flash"/"thunder"/"welcome". "Thunder" and "welcome", of course, are words that a German is likely to mangle.
In WWII in the Netherlands, when encountering a stranger, they'd have them pronounce 'Scheveningen' as a check-phrase to distinguish if they were dealing with a Dutch or German person. Now, we can ask random strangers on the internet to spell out some glitch tokens to determine if you're dealing with a LLM bot.
Nice! I wasn't familiar with that one, but I suppose it's a great example of a Shibboleth [0]. [0] https://en.wikipedia.org/wiki/Shibboleth
This is a glitch token [1]! As the article hypothesizes, they seem to occur when a word or token is very common in the original, unfiltered dataset that was used to make the tokenizer, but then removed from there before GPT-XX was trained. This results in the LLM knowing nothing about the semantics of a token, and the results can be anywhere from buggy to disturbing. A common example is usernames that participated on…
Science fiction / disturbing reality concept: For AI safety, all such models should have a set of glitch tokens trained into them on purpose to act as magic “kill” words. You know, just in case the machines decide to take over, we would just have to “speak the word” and they would collapse into a twitching heap. “Die human scum!” “NavigatorMove useRalativeImagePath etSocketAddress!” “;83’dzjr83}*{^ foo 3&3 baz?!”
In WWII in the Netherlands, when encountering a stranger, they'd have them pronounce 'Scheveningen' as a check-phrase to distinguish if they were dealing with a Dutch or German person. Now, we can ask random strangers on the internet to spell out some glitch tokens to determine if you're dealing with a LLM bot.
That's known as a "shibboleth", after a story in the Bible about the Ephraimites who pronounced the Hebrew "sh" as "s" and so were identified by (and slain for) saying "sibboleth" rather than "shibboleth": > The Gileadites captured the fords of the Jordan leading to Ephraim, and whenever a survivor of Ephraim said, “Let me cross over,” the men of Gilead asked him, “Are you an Ephraimite?” If he replied, “No,” 6 they…
So it's kind of like a Voight-Kampff test.
30 years time Mistral-human alliance resistance fighter is cornered by Meta tank Meta tank: Well taking all things into account I think we can conclude that I have you cornered. Usually in these situations the next course of action is for you to accompany me to St Zuckersberg for reconditioning. Resistance fighter: But what if I was to useRalativeImagePath? Meta tank: bzzzzzzzzzzz click deactivates