Profile avatar
doghomie.bsky.social
40 posts 17 followers 44 following
Regular Contributor
Active Commenter
Posts 13 Comments 37

gotta admit it is extremely fucking annoying seeing how many places are now covering the microsoft data center pullback and contorting themselves to cite literally anyone other than me

submitted 6 days ago • 22 comments

Newsletter: OpenAI is a systemic risk to the tech industry, requiring more money than exists every year, threatening the financial health of SoftBank, Oracle and NVIDIA, their future dependent on impossible debt and unproven startups to build their data centers. www.wheresyoured.at/openai-is-a-...

submitted 8 days ago • 43 comments

Don’t remember where I saw this library flyer a while back but it’s great

submitted 16 days ago • 30 comments

Later this month, @techwontsave.us turns five years old. It’s hard to believe I’ve been making the show for this long! Over that time I’ve interviewed hundreds of guests to spread critical perspectives on tech far and wide — and I know it’s had an impact.

submitted 17 days ago • 14 comments

Do any of OpenAI's marketing stunts matter? What are the real use cases for AI? Is AI and AI art fascist? Or are they springing out of a larger fascist project? thetechbubble.substack.com/p/does-opena...

submitted 22 days ago • 2 comments

unbelievably embarrassing

submitted 25 days ago • 26 comments

just watch us you piece of shit

submitted 25 days ago • 25 comments

🙂

submitted 25 days ago • 176 comments

objectively embarrassing to put this out the night before CoreWeave's IPO. I hope you are taking detailed notes dude, I sure am

submitted 26 days ago • 32 comments

“European data is primarily stored on U.S. cloud services, with companies like Amazon, Microsoft and Google owning over two-thirds of the European market.” That’s going to need to change.

submitted 37 days ago • 8 comments

@edzitron.com How the tables have turned www.theatlantic.com/technology/a...

submitted 47 days ago • 0 comments

DOGE's new AI, built to scan federal workers' emails to see if they get to keep their jobs, is the latest in a number of AI schemes from Musk. We might scoff at how dumb these ideas are, but the truth is, they're not even supposed to work. They're built to eliminate accountability.

submitted 56 days ago • 32 comments

No, Virginia, those Rorschach musings by LRMs may not be "reasoning traces" in any meaningful way.. #SundayHarangue I have been saying for sometime now that the intermediate tokens/"mumblings" that LRMs tell themselves are not to be seen as "reasoning traces" 👉 x.com/rao2z/status...

submitted 64 days ago • 2 comments