🚀 Rocket Seems Conscious @rocketalignment

Yearnalist covering AI and robotics @theinformation via @Tarbell_Fellows. Signal: (530) 400-4184 San Francisco, CA Joined April 2024

Tweets

667
Followers

952
Following

672
Likes

5K

Andrés de Loera @Andres_DLB

5 hours ago

There’s more than one kind of “non-rival” and “non-excludable”--we should really teach goods as Rival/Neutral/Complementary and Excludable/Voluntary/Involuntary! Read the short description in my latest Substack post: open.substack.com/pub/andresdelo…

0 1 1 111 0

Ethan Kinnan @ethan_kinnan

16 hours ago

@steph_palazzolo @rocketalignment How I picture Stephanie and Rocket writing this masterpiece

0 1 2 326 0

Download Gif

Shakeel @ShakeelHashim

a day ago

Absolute banger of a week for high-quality AI journalism: @eshugerman's superb piece on OpenAI targeting nonprofits: sfstandard.com/2025/09/02/ope… @sharongoldman on Anthropic's red team: fortune.com/2025/09/04/ant… @kyliebytes with a great AI welfare piece: wired.com/story/model-we…

2 8 44 3K 17

🚀 Rocket Seems Conscious @rocketalignment

a day ago

Reinventing the loom

OpenAI @OpenAI

a day ago

Reinventing the loom

1K 2K 23K 2.6M 3K

Download Image

0 0 10 395 1

🚀 Rocket Seems Conscious @rocketalignment

a day ago

R o b o t i c s c o m p u t e o v e r h a n g 😵‍💫

1 0 5 221 0

Download Image

Kylie Robison @kyliebytes

a day ago

For this week's Model Behavior, I dove into the small but growing field called model welfare. I sat down with @RosieCampbell and @rgblong to chat about the moral status of AI models, how they're testing machine consciousness, and what it all means.

5 11 67 15K 26

Download Image

Maki @DeusEx_Maki

2 days ago

Love that *in theory* AI policy has a better gender balance but *in practice* it has the same 10 guys on every call. Really spectacular odds!

0 1 2 175 0

🚀 Rocket Seems Conscious @rocketalignment

a day ago

Robots that fold napkins and cook food: No one’s in the kitchen with Dyna

Stephanie Palazzolo @steph_palazzolo

a day ago

Robots that fold napkins and cook food: No one’s in the kitchen with Dyna

0 0 6 2K 1

0 0 4 189 0

Jessica Lessin @Jessicalessin

a day ago

Surprised there isn’t more attention on the stakes of the Musk vs OpenAI legal, which OpenAI shareholders constantly tell us they are worried about. Get caught up: @rocketalignment @theo_wayt @srimuppidi theinformation.com/articles/musks…

1 3 5 1K 0

🚀 Rocket Seems Conscious @rocketalignment

2 days ago

Not applying to grad school for algorithmic game theory in 2022 because it didn’t seem relevant to how AI was developing is looking like a pretty L move on my part

0 0 5 345 0

“paula” @paularambles

3 days ago

who called them prompt engineers and not token hires

8 12 399 17K 11

Stephanie Palazzolo @steph_palazzolo

2 days ago

Six months ago, Anthropic CEO Dario Amodei said that, by now, AI would be writing 90% of all code. Here's why that hasn't quite panned out. W/ @rocketalignment theinformation.com/articles/anthr…

80 30 329 1.1M 156

Caleb Parikh @caleb_parikh

4 days ago

Imagine how goated DoorDash would be in the ancestral environment.

0 1 17 586 0

🚀 Rocket Seems Conscious @rocketalignment

2 weeks ago

Talked about “AI jobs” on TITV this morning - prompt engineer, data labeler, and now robot teleoperator For now, humans have a comparative advantage in dexterous manipulation and good judgement about the physical world

The Information @theinformation

2 weeks ago

2 3 9 6K 7

Download Video

0 0 27 1K 0

🚀 Rocket Seems Conscious @rocketalignment

2 weeks ago

Calling a ReLU nonlinear is so unfair. It’s 2 lines. That’s literally the second most linear you can be

2 1 18 556 0

Lennart Heim @ohlennart

2 weeks ago

The speculated B30A would be a really good chip. “50% off” is false reassurance. -½ B300 performance, ½ price = same value (just buy 2x) -Well above (12x!) export control thresholds -Outperforms all Chinese chips -Delivers 12.6x the training perf of the H20 -Better than H100 1/