Hacker Newsnew | past | comments | ask | show | jobs | submit | arisAlexis's commentslogin

Amazon went all in robotics they have their ai training chips, they build their own satellite starlinks and run the one if the 3 clouds.

Like he understands tech

Only it's much more exponential

Totally wrong. He has been voicing this for ages and for specific reasons.

How do you know the price of a unit ?

I remembered $1m from when I was in their booth at SC24, but when I just looked, I was wrong. It is worse...

https://www.datacenterdynamics.com/en/news/cerebras-unveils-...


You have no idea of the value this brings. Asml machines cost dozens of times more. So?

Ok, tell me the value.

They were afraid for the untested tech but it looks like a leap in speed now

This is nonsense what do you mean? Mistral uses Cerebras for their LLMs as well. [0]

It's certainly not "untested".

[0] https://www.cerebras.ai/blog/mistral-le-chat


Tested at Mistral’s scale is a very different thing to tested at OpenAI’s scale.

The scale of being "tested" clearly convinced Meta (beyond OpenAI's scale) [0] HuggingFace [1], Perplexity [2] and unsuprisingly many others in the AI industry [3] that require more compute than GPUs can deliver.

So labelling it "untested" even at Meta's scale as a customer (which exceeds OpenAI's scale) is quiet nonsensical and frankly an uninformed take.

[0] https://www.cerebras.ai/customer-spotlights/meta

[1] https://www.cerebras.ai/news/hugging-face-partners-with-cere...

[2] https://www.cerebras.ai/press-release/cerebras-powers-perple...

[3] https://www.cerebras.ai/customer-spotlights


Meta didn't offer it. They offered the free llama version on their cloud. Maybe now Zuck will be conincrto buy their chips though

Denial everyone. Amazon will have the same profits running on AI and robots with minimal expenses. All the other companies will follow. Wake up to reality.


There is widespread denialism in the tech community about this. Most think they will never get replaced until they get a call from the manager. It's called normalcy bias.


In the 90s, I remember explaining to my college econ professor that the cost of developing software would decrease to zero within 30 years. He kind of got it but then kept encouraging kids to study computer science for the money.


Kids in the 90's would have had long fruitful careers making much more than the median american for relatively easy and stable work.


I saw the writing on the wall the first day in the office. Microsoft was already dumping loads of R&D money into software development automation. They were very motivated. Everyone hated software developers because we made too much money relative to people with important positions. And, we were too weird. To them, we were 'blue collar'. I figured with exponential progress it wouldn't take long. Pretty much my strategy after college was bank and invest and get out as early as possible then go hack on my own projects until my end of days.


HN has always been like this though. Smart but lots of hubris. They believe the world will always value their current skills as much as pre-Covid


Why hinge it on COVID?


The fact that Tedros is still the head while he mishandled completely the biggest catastrophe of our era should be taken seriously.


Forget the inevitable singularity and start the pseudoscience of technical analysis for stocks. Smart.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: