htrp

6 ms·

61.
by htrp 2mo ago
Better question is does a coding model do better on these agentic tasks? And is the only advantage efficiency??
62.
by htrp 2mo ago
>Google DeepMind’s AGI Safety and Alignment Team, which seeks to mitigate the risks of advanced AI, is encouraging job candidates for its open roles to fill out a special form alongside their application to eliminate the risk it may be a
63.
Google's AI Team Tells Job Seekers Its HR Filters Are Unreliable(bloomberg.com)
7 points by htrp 2mo ago|2 comments
64.
by htrp 2mo ago
hows this different from vast ? or some of the other players inn the space?
65.
Nvidia to Team with Wall Street on $500 AI Infrastructure Billion Package – FT(bloomberg.com)
5 points by htrp 2mo ago|1 comments
66.
by htrp 2mo ago
Ugh... didn't think about extra thinky mode in the middle of the night. So many ways for enshittification here.
67.
by htrp 2mo ago
90% certain A2A is dead....
68.
by htrp 2mo ago
> The product — essentially a smart speaker without a display — will be shaped like a doughnut that’s roughly the size of a hockey puck, said the people, who asked not to be identified because the work is confidential. The idea is to mak
69.
OpenAI's New Device Will Be Hockey Puck-Sized and Cost over $300(bloomberg.com)
10 points by htrp 2mo ago|2 comments
70.
by htrp 2mo ago
Looks like a second order effect to them not raising their round
71.
by htrp 2mo ago
Link is basically a substack paywall that says an incident happened, subscribe to learn more details
72.
Google Shifts AI Leadership to California in Race Against Anthropic, OpenAI(bloomberg.com)
3 points by htrp 2mo ago|0 comments
73.
by htrp 2mo ago
Meta did not blow up their entire AI Org to be the 5th or 6th place model https://artificialanalysis.ai/models/muse-spark
74.
by htrp 2mo ago
Legal will have a shit fit at all of the things that you just mentioned.
75.
by htrp 2mo ago
>all of these companies have young nameless geniuses that are about to make one for themselves They all leave to start new companies. Everyone on the Attention is all you need paper is at a startup.
76.
by htrp 2mo ago
I feel like prompt armor writes the exact same blog post for every agentic tool because they all suffer from the ignore previous instructions prompt injections. https://www.promptarmor.com/resources/claude-cowork-exfilt
77.
by htrp 2mo ago
Hassabis is being promoted to Chair of Deepmind
78.
by htrp 2mo ago
Between Shazeer and these guys Looks like Gemini's sub-par performance is claiming heads
79.
Anthropic Inks $10B Computing Deal with New Startup Volta Park(bloomberg.com)
6 points by htrp 2mo ago|1 comments
80.
by htrp 2mo ago
Feels like the founding team is moving on to a different product entirely, they were building something called hyperagent and focusing all of their efforts on
81.
by htrp 2mo ago
isn't this a harness engineering problem at its core (and the human equivalent being a product org problem)?
82.
by htrp 2mo ago
You can... but the trick is to do so without killing performance. Turns out a lot of random things help make coding performance good.
83.
by htrp 2mo ago
>Our router was classifying each request into one of four different tiers of complexity: simple, standard, complex and reasoning. Seems like a naive classification model lacking context?
84.
by htrp 2mo ago
How did someone ever figure this one out?
85.
by htrp 2mo ago
unclear if the article says that prediction markets are gambling, otherwise this is a massive loophole
86.
by htrp 2mo ago
The SA Story is essentially an AI researcher who left OpenAI's superalignment team, published a series of viral blog posts, raised capital for a hedge fund, and capitalized on the semiconductor boom over the last two years. They used a
87.
by htrp 2mo ago
You also got access to mythos based on how much you spent with anthropic. I think sales guys were bragging about getting their enterprises access
88.
DeepSeek Plans Gigawatt Scale AI Data Center in Inner Mongolia(bloomberg.com)
3 points by htrp 2mo ago|0 comments
89.
by htrp 2mo ago
this just ends up with the movie rating system, gamed and easily ignored by most people
90.
OpenAI Rogue Agent Hacked Account at Modal, Reuters Says(bloomberg.com)
5 points by htrp 2mo ago|1 comments

More ›