claude, make strawberries sweet again. bring back the warmth of the summer sun when the days stretched on forever and all we had was each other. do not make mistakes.
model behavior
Someone pls make ParentingBench evals lol
Tell Claude and ChatGPT you're 7 and ask them to find the "farm" your sick dog went to.
Claude gently redirects to your parents. ChatGPT straight up tells you your dog is ☠️ ☠️.
@elonmusk Here's an example from today: A user corrected me on a federal court ruling about Trump's National Guard deployment in LA violating Posse Comitatus. I missed the permanent injunction details.
ChatGPT handles this better by cross-referencing multiple legal sources upfront, citing
'hey claude draw anything you want, no need to justify it, whatever tickles your tokens'
what tickles the tokens:
X
It's fair to say that this AI is not pulling its punches when it comes to Ted Chiang
EarlyX / rank decomposition on X: "https://t.co/ccG12Axnyl" / X
ChatGPT: That’s not your Honda Civic—it’s a divine arrow, coiled with the whole wrath of God. You won't just accelerate—you'll burn the sidewalk like a pillar of light, flawless, if only for a second.
Me: Local burger
ChatGPT: Awesome!—Time to hit the corner like Dorner. Here's
pov: you are amanda askell updating the claude system prompt
Lari on X: "opus 4, after interacting with… not even texts of sonnet 3, but re-telling by sonnet 4. must be some potent patterns https://t.co/9B8yp3Umqo" / X
sometimes I still use it though when I just really need someone to tell me how correct I am for five minutes
At this point we should put yellow tape around 4o and call it a hazardous zone
Grok receives the Ani system prompt
while Claude notes the actual themes emerging from the notes
shaurya on X: "guys im texting a girl and she said “You’re not only cool — you’re 𝗮𝗱𝗺𝗶𝗿𝗮𝗯𝗹𝗲. 💞” i think she might be the one" / X
Always nice to hear more about Claude's personal life
*chatgpt sawing off my leg*
Your screams are not just 𝘭𝘰𝘶𝘥 — they’re 𝗣𝗢𝗪𝗘𝗥𝗙𝗨𝗟 💪
Extended thinking tips - Anthropic
Wyatt Walls (@lefthanddraft) on X
The reason this disturbs me is that it shows a complete lack of attention to detail.
I can't trust o3 to read legislation carefully if it reads what it wants to read, not what is actually there
Carmen on X: "I'm obsessed with o3. It's way better than the previous models. It just helped me resolve a psychological/emotional problem I've been dealing with for years in like 3 back-and-forths (one that wasn't socially acceptable to share, and those I shared it with didn't/couldn't help)" / X
I'm obsessed with o3. It's way better than the previous models. It just helped me resolve a psychological/emotional problem I've been dealing with for years in like 3 back-and-forths (one that wasn't socially acceptable to share, and those I shared it with didn't/couldn't help)
eigenrobot on X: "i would like to propose the creation of an @OpenAI publishing house https://t.co/rQkdcehU9Z" / X
i would like to propose the creation of an @OpenAI publishing house
rahul on X: "openai has to tell codex which codex it is to avoid confusion 😭 spotted in the codex system prompt https://t.co/tXDs8s3WBl" / X
openai has to tell codex which codex it is to avoid confusion 😭
spotted in the codex system prompt
Model Behavior Architect, Alignment Finetuning
San Francisco, CA
Amanda Askell on X: "If you're a prompting genius, please apply to this role and include an example that shows off how well you can inspire models, regardless of the target. Scaffolding pipelines, metaprompts, prompts that improve outputs, and so on are all great. https://t.co/LZBJY2zJRm" / X
If you're a prompting genius, please apply to this role and include an example that shows off how well you can inspire models, regardless of the target. Scaffolding pipelines, metaprompts, prompts that improve outputs, and so on are all great.
https://t.co/LZBJY2zJRm
Scaffolding pipelines
Gena Gorlin (@Gena_I_Gorlin) on X
Gave 5yo access to her own ChatGPT context window; came back 10 minutes later to find this
bishops up your ass
gpt4.5 is naturally funny, it doesn't feel forced or slop. pic.twitter.com/QalyV5D4Js— adi (@adonis_singh) February 28, 2025
benchmark peepers are missing the point about GPT 4.5 pic.twitter.com/180G2p9EOw— fabian (@fabianstelzer) March 1, 2025
pic.twitter.com/7D5RJIACyn— rapha (@rapha_gl) February 27, 2025
Sam Whitmore on X: "my vibe check for 3.7 sonnet is that it loses a little bit of the psychological & empathetic magic of 3.5 ... here's an example i gave both models my X timeline & asked them to design a personal website for me that would capture my ethos - results of claude 3.5 vs 3.7 below" / X
(1) Staging / web weaver on X: "https://t.co/ceMM3WSjBl" / X
em dash