To be fair, you don’t need an LLM for that. The earliest text-based RPGs operated on free text input from the player, which would trigger actions based on keywords. (e.g., “go north”, “pick up lantern”, etc.")

You could do the same thing with dialogue. It’s just a bunch of work and kinda clunky, but so is talking to a chatbot.

I had one of these last week. Refreshed the page and got a standard image-based captcha. It won’t be long before this QR/device attestation shit becomes the norm though.

Where is Maduro?

I saw the clip. He did say no. Then the dude asking says something like “then why are we here?” and I wholeheartedly agree.

Google’s been doing that for ages. The thing that finally got me to pull the trigger on GrapheneOS last year was my Pixel telling me I had “over 8GB of memes” on my phone and asked if I wanted to clean up my storage.