Reasoning models on a simple yes/no question:

GPT 6 Luna when I tell it good morning.

😅

Accurate 😂

😁😁

This is Qwen. But you see it reach for the poison, stops itself at the last second, and somehow chooses the right goblet in the end.

Asked yes or no and somehow we’re already at the tavern getting the quest briefing 😅

😂

😂

Should be an advert for system one models

This is every Group meeting I hold in @bot ....after I accidently ask how everyone is doing this morning out of habit.

Life before Jev

A "truly dizzying intellect" would dump out both cups, rinse them out (with wine), and pour fresh wine

Zorro

I was the great fool when I visited my bribed and corrupted family for Christmas dinner in 1996 — as many other have been great fooled.

99% of ceos don't understand the oxymoron that is a probabilistic agent. the reasoning models at least are true to their nature lol, poised to show off the "intelligence" they ended up just showing the "probability", which really puts the oxymoron in full display

LMAO

And that's the crux of the matter. If it's not in one cup, it's in the other. Here's where i draw the line, i won't be drinking from either cup. blah blah more metaphors...

👌😂

Wine neutralises the poison, drink up!

Is this the show where both cups were poisoned but the masked dude was immune to it? Or is this a parody, 😂