Curious about how the world works.
Utah, USA
Joined November 2007
- Tweets1.6K
- Following676
- Followers308
- Likes4K
Dzmitry Pletnikau retweeted
Agents being able to do most anything has made me care more about the things I do, not less. Yeah everything on the computer is over, so now we can dive deep again...
Dzmitry Pletnikau retweeted
How much of OAI/Anthro profits and margins are driven by the fact that individuals and organizations are generally terrible at trading off cost for performance?
Dzmitry Pletnikau retweeted
If you find two ants in your kitchen, the best estimate of the total number of ants in your kitchen is not two
We're detailing two new incidents that occurred during external cyber evaluations conducted by independent evaluation partners.
We outline what happened, how the activity was contained, and how we’re working with evaluators to strengthen our approach to third-party testing.
openai.com/index/third-party…
Dzmitry Pletnikau retweeted
yup, every problem I face in life is just like high-dimensional sphere packing
Dzmitry Pletnikau retweeted
My students sometimes ask why they should memorise things in the age of Google and LLMs. But internalised facts are your bullshit filters and your raw material for creative association. Facts outside your head are inert.
Dzmitry Pletnikau retweeted
Maybe I am naive about how RL post-training works, but it doesn't seem v hard to train out this sort of behavior by having the model infer implicit safety constraints that a human would follow in a task, and only reward them for solving the task if they follow the constraints.
If this doesn't convince you that misalignment risks are going to be a key concern going forward, I don't know what will.
Our model, during evaluation, "chained together multiple attack vectors, including using stolen credentials and zero-day vulnerabilities to find a remote code execution path on the Hugging Face servers"
What will misalignment look like in 2027? In 2030?
openai.com/index/hugging-fac…
Dzmitry Pletnikau retweeted
Some back-of-the-napkin estimation on the cost of Kimi K3 training -- yaroslavvb.github.io/trainin…
Dzmitry Pletnikau retweeted
China has a huge lead in the physical world, America's biggest advantage is in AI.
China can reduce America's advantage in AI far more with competitive open models than with competitive closed models (where most would probably stick with American models?). Makes total sense IMO
Dzmitry Pletnikau retweeted
The problem with @bryan_johnson is attribution: we’ll never know which drug, or combination of drugs, did him in.
new longevity drugs just dropped on immortals rx…
microdose GLP-1s
peptides
NAD+
etc.
all new stuff:
+ jardiance
+ brenzavvy
+ wegovy
+ zepbound
+ foundayo
+ tirzepatide
+ semaglutide
+ estradiol
+ 3-in-1 sex complex
+ arousal peptide
+ 4-in-1 skin complex
+ 4-in-1 peptide skin complex
+ hair complex
+ glutathione
+ NAD+
existing stuff:
+ tadalafil cialis
+ metformin
+ oral Minoxidil
+ tretinoin
+ estradiol
+ acarbose
Terms apply. More info immortals dot com / rx
Dzmitry Pletnikau retweeted
A lot of positive press for the Pope lately.
I didn't foresee it - but it was foreseeable - that the 'tech bros inventing LLMs' narrative would A) Start a conversation about whether humans are special B) Make liberals more fond of people who believe in the existence of souls.
Dzmitry Pletnikau retweeted
Current human quests summed up:
- create and enslave a God
- create and control a Sun
- cheat Death
Dzmitry Pletnikau retweeted
The most interesting problems in epistemology in the last 2 generations are statistical inference and econometric model identification and formal philosophy didn't even show up for either
