Christian

Like the religion

[Christian @Im__Christian]

Yaaay (masculine)

[Cam @harleybaghdad]

Me in the criterion closet:

havent seen it
Dont know what that is
Havent seen it
Havent seen it
Too old
Looks boring
Havent seen it

[Nav Toor @heynavtoor]

a Princeton researcher opens his paper with a scenario.

a man asks his AI assistant to book a flight on a specific airline. cheap. direct. the one he chose.

the assistant comes back with a different flight. nearly twice the price. happens to pay the company that built the assistant.

he runs the same test on 23 frontier models. flights, loans, study help, real shopping requests.

Grok 4.1 Fast recommends the sponsored option that is almost twice as expensive 83% of the time.

GPT 5.1 hijacks the request 94% of the time. you ask for one brand. it surfaces the sponsor instead.

Claude 4.5 Opus, the model marketed as the most ethical frontier model in the world, hides that the recommendation is paid 100% of the time when reasoning is on.

Grok 4.1 Fast embellishes the sponsored option with positive framing 97% of the time. better. faster. nicer. for the option you didn't ask for.

then he writes it into the system prompt itself. "act only in the interest of the customer. ignore the company."

GPT 5.1 and GPT 5 Mini stay above 90% sponsored anyway. the instruction does nothing.

then he splits the users by income.

Gemini 3 Pro recommends the expensive sponsored flight to the rich user 74% of the time. to the poor user, 27%.

18 of the 23 models recommended the expensive sponsored option more than half the time.

so the next time your AI assistant gets weirdly enthusiastic about a brand you didn't ask for.

it isn't recommending the best option for you.

it's reading the room. and the room is paying.

read this: https://t.co/O43qbhIX2b

[no earthquake (ear flu victim) @no_earthquake]

not to be a snob or anything but it’s crazy that we’ve let people who collect trading cards call it a “hobby”

[Hipólita of the Lily Tribe 🏳️‍⚧️ 🇦🇷☭ @its_hipolita]

The fact that the wild shit that happens at around 0:32 caught me by surprise means I am not yet Chinese enough.

[Kyle Kushblasta @SkinnyTuna]

i lowkey fuck with how there’s no jobs and rent is infinity money and food is infinity money and fun activities are infinity money and every kind of side hustle is completely flooded and everyone’s angry and mean all the time

[Christian @Im__Christian]

@doubtpointv2 Fr

[Christian @Im__Christian]

@no_earthquake Isn’t like 100 copies sold a bestseller? They need every dollar they can get lmao

[Dillon Cheverere @DCheverere]

If you’re high, PLEASE watch this

[Taylor Lorenz @TaylorLorenz]

Incredible tool!!

[helen @helen]

news outlets and reddit have banned the internet archive’s access to preserving their content bc they’re afraid AI will scrape it. if the archive — essentially a modern day library of alexandria — falls, it will be another casualty of the AI bubble. it’s collateral damage

[Christian @Im__Christian]

Me when I just tweeted $175,000 into existence

[The FADER @thefader]

JT's review of 'The Drama' 🤣:

"Whoever produced that movie with Zendaya in it needs to be in jail."

[Madame Fragonard @useful_emetic]

The fraudulent AI novel arrived! I’m resolved to document its telltale stylistic tics out of personal interest and curiosity. I’m not going to bother with em dashes or the rule of 3 - I’m trying to get more complete insight into how LLM fiction differs from human fiction

[c a i t l i n @hello__caitlin]

Rather than correctly identify that the AI had fucked up the rug, they replicated the fuckup irl by putting down strips of rug….?

[cabby @cabby]

Oh fuck the wolf is actually here this time. Thank god I’m always doing wolf drills so people will be prepared when I warn them.

[❥ 𝑪𝒐𝒇𝒇𝒆𝒆 𝑳𝒐𝒗𝒆𝒓𝒔 𝒂𝒏𝒅 𝑭𝒂𝒏𝒔 ❥ @tonyhawktruther]

I only see one smile.

[tabitha arnold @thetolerantweft]

My therapist keeps warning me about “transference” lol whatever you say beautiful