Posted on 12/09/2024 9:05:03 AM PST by SeekAndFind
Amazon’s cloud computing arm Amazon Web Services Tuesday announced plans for an “Ultracluster,” a massive AI supercomputer made up of hundreds of thousands of its homegrown Trainium chips, as well as a new server, the latest efforts by its AI chip design lab based in Austin, Texas.
The chip cluster will be used by the AI startup Anthropic, in which the retail and cloud-computing giant recently invested an additional $4 billion. The cluster, called Project Rainier, will be located in the U.S. When ready in 2025, it will be one of the largest in the world for training AI models, according to Dave Brown, Amazon Web Services’ vice president of compute and networking services.
Amazon Web Services also announced a new server called Ultraserver, made up of 64 of its own interconnected chips, at its annual re:Invent conference in Las Vegas Tuesday.
Additionally, AWS on Tuesday unveiled Apple as one of its newest chip customers.
Combined, Tuesday’s announcements underscore AWS’s commitment to Trainium, the in-house-designed silicon the company is positioning as a viable alternative to the graphics processing units, or GPUs, sold by chip giant Nvidia.
The market for AI semiconductors was an estimated $117.5 billion in 2024, and will reach an expected $193.3 billion by the end of 2027, according to research firm International Data Corp. Nvidia commands about 95% of the market for AI chips, according to IDC’s December research.
“Today, there’s really only one choice on the GPU side, and it’s just Nvidia,” said Matt Garman, chief executive of Amazon Web Services. “We think that customers would appreciate having multiple choices.”
(Excerpt) Read more at wsj.com ...
A key part of Amazon’s AI strategy is to update its custom silicon so that it can not only bring down the costs of AI for its business customers, but also give the company more control over its supply chain.
That could also make AWS less reliant on Nvidia, one of its closest partners, whose GPUs the company makes available for customers to rent on its cloud platform.
Off-topic, but speaking of Amazon and AI: I’ve noticed that now, if you ask a question about an item, instead of being answered by a person who bought and reviewed it you just get an answer compiled from reviews by an AI.
I don’t like that.
I am sick of hearing about AI.
Musk’s X.ai recently announced the world’s largest super computer working out of memphis tennessee. I’m reading it has broken all kinds of records. Its showing that the scaling laws can go to the moon and back.
The big thing is its ability to enable coherence at scales all the other groups working on AI have not been able to attain. Coherence means the ability of every chip to talk to every other chip simultaneously. The limit is 30 (measurement). X.ai achieved a limit of 100(measurement) with a theoretical limit of 1 million (measurement).
I hardly know what I’m talking about. It looks bigger than the Amazon announcement, except if Amazon can make chips at a scale that is as good as Nvidia and can work interoperably with Nvidia Cuda software.
Now if they could only find someone who could write a decent search engine for their website. Trying to find a specific item on Amazon is as annoying as trying to do so at Costco. And probably for the same reason.
Commercial chips have embedded malware, exploits and back doors that can be used to spy on systems , take remote control of them or take them down.
It's really hard to detect these malicious codes because they are implemented at the circuit level and they are part of the silicone circuit architecture. Chip making is becoming a generic technology due to AI circuit design and automated chip processing so it is well within the grasp of a company the size of Apple, Amazon, Google or GM . This is especially true of specialized, dedicated purpose chips like GPU processors
“I am sick of hearing about AI.”
Pick One:
Why people are sick of hearing about ai
ChatGPT said:
ChatGPT
The growing backlash against AI discussions stems from several factors, including concerns over its rapid rise, its impact on jobs, the ethical questions it raises, and the sheer volume of conversations around it. Here are some of the key reasons why people might be “sick” of hearing about AI:
1. AI Fatigue
With AI’s increasing presence in media, marketing, and everyday conversations, many people feel overwhelmed. It’s constantly in the news, from its implications in the job market to its role in creative industries, leading to a sense of saturation. The sheer frequency of AI-related topics can make people tune out or feel tired of hearing about it.
2. Fear of the Unknown
AI has advanced so quickly that it’s difficult for many to fully grasp its capabilities, leading to uncertainty and fear. People may be exhausted by the constant debate over how AI will affect their lives—whether it’s concerns about job displacement, surveillance, or even the fear of an AI “takeover.” These anxieties can make people avoid discussions about it.
3. Hype vs. Reality
There’s often a gap between the promises made about AI and its actual, practical applications. While the media may hype AI as a solution to everything from curing diseases to solving climate change, the reality is more complex. Many people are frustrated by the overblown claims and want more grounded discussions about AI’s real potential and limits.
4. Job Insecurity
AI’s potential to replace human workers in fields like customer service, manufacturing, and even creative industries has led to a growing sense of insecurity. People worry about their livelihoods being replaced by automation, and as discussions about AI’s impact on employment continue, they may grow tired of hearing about it.
5. Lack of Control and Regulation
Many are frustrated by the lack of regulation surrounding AI development. It feels like AI is evolving faster than laws or ethical frameworks can keep up, leaving people uneasy about its unchecked power. The absence of meaningful oversight can make discussions about AI seem more like a technological runaway train than a thoughtful exploration of its benefits and risks.
6. Overuse in Marketing
AI has become a buzzword used in everything from advertisements to product descriptions. Companies claim their products are “AI-powered” even when the term is applied loosely or inaccurately. This “AI-washing” leads to confusion and skepticism, making people weary of hearing about AI in marketing contexts.
7. Ethical Dilemmas
Discussions around the ethical implications of AI—like data privacy, surveillance, bias, and decision-making—are often fraught with tension. These discussions can feel morally exhausting, especially when solutions seem far off or unattainable, leaving people feeling disillusioned.
8. Echo Chamber of Opinions
AI discussions often happen within echo chambers, where opinions are polarized. On one side, you have the tech enthusiasts who see AI as the future, and on the other side, you have critics who fear its negative effects. The resulting back-and-forth can be exhausting, especially when it feels like no consensus or meaningful action is being taken.
9. Cultural Shift
As AI increasingly affects daily life, it forces people to rethink how they interact with technology, work, and even their identity. This cultural shift can be unsettling. People don’t just dislike hearing about AI; they dislike how it’s forcing them to confront deeper questions about their lives in a way they don’t feel ready for.
In essence, the sheer ubiquity of AI discussions—combined with fears, ethical concerns, and unmet expectations—has created a perfect storm of AI fatigue. People aren’t necessarily against AI itself, but they are frustrated with the constant bombardment and the lack of meaningful, grounded conversations around it.
“and can work interoperably with Nvidia Cuda software”
Most people aren’t aware of how important their software is!
AI: “solving climate change”
Yup—AI is still stuck on stupid.
Once it rejects “climate change” as leftist propaganda then we can start to worry about real artificial intelligence that can separate creepy propaganda from verifiable facts.
Why did you excerpt? To mislead by eliminating the context?
Any use of the term “climate change” by AI that does not include clear condemnation of the term was correct context.
AI was slurping up the slop.
“Any use of the term “climate change” by AI that does not include clear condemnation of the term was correct context.
AI was slurping up the slop.”
Context: The sentence referred to hype.
There are a lot of Bay Area IT salespeople who need to keep up their lifestyles, by overhyping and overselling technology, don't you care about their status and social well being ?
Ironically, these are the same people who are constantly virtue signaling about Climate Change ....
Kill AI.
The world is so crazy, everything up is down....clot shot, lawfare, globull warning,
The unibomber makes sense? Krazy.
Is Greta not so krazy? Wait a minute....
“Why people are sick of hearing about ai
ChatGPT said:”
That’s a great self-analysis by AI - more honest than most humans
More proof that it’s not just a fad and we better learn to deal with it.
Kill AI.
AI is the “Image of the Beast.”
AI is the Anti-Christ.
Most people aren’t aware of how important their software is!
.........
The extent of my understanding is that I know how to say Nvidia Cuda software and that the moat around Nvidia GPU’s includes the Cuda software.
The really big stuff from Nvidia—on the scale of the X.ai third generation announcement—is the announcement last summer that Nvidia and Dell had teamed up to create AI factories.
Ai Factories are basically Elon Musk in a box. I did some research on this. They enable the creation of digital twins of any wigit, tool, machine, machine that makes the machine and factory floor on which all the production takes place.
These thing create simulation environments that obey all the laws of physics. So when a manufacturer want to create a new product that will perform to specs—they can run simulations until they have a simulation that peforms to spec and then and only then do they build it. Once they have the wigit,tool,or machine—then the manufacturer can use the AI factory digital twin—ie the simulation that obeys all laws of physics—to create the machine that makes the machine, wigit or tool. They keep simulating until they get a machine that makes the machine that meets the specs. Once they have the product they want to build and the machines that will build the product—then they can use the AI factory to design a real factory. They run simuations on that until all the pieces fit together optimally. Then they create the real factory. Once the factory is up and rolling then they feed the AI factory real time data from the real factory. So that they can continuously optimize the product and the manufacturing process.
When this thing gets out at scale—basically the whole of the USA will be able to move at Musk speed because he’s been able to figure out how to do this on his own for his inhouse work.
“don’t you care about their status and social well being ?”
Yes, yes, I’m sorry. I was being cruel and thoughtless. I care deeply about India.
Yep, it is another slide backwards for the site, and a big one.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.