#artificial intelligence chip

[manmishra]

🚀💻 BREAKING: Nvidia’s AI Ultra Chip is coming! 🤯🔥 With Reuben Architecture, this next-gen chip will make AI computing ⚙️💡 faster, smarter & more powerful! 🤖⚡ 💊🏥 Healthcare: Faster diagnosis & drug discovery 💉🧠 🚗🚦 Self-driving cars: Real-time decision making 🏎️🔋 💰📊 Finance: Accurate predictions 💹📈 🎮🕹️ Gaming: Ultra-realistic graphics 🎨👾 But can Nvidia stay ahead of AMD, Intel, Google & Microsoft? 🤔💡 👉 Read more! 📖🔍 #NvidiaAI #FutureOfAI #TechNews #AIUltraChip 🚀💻

[rotting-brains]

they couldn't do a mcsm season 3 because if petra had to suffer for plot reasons for a single other episode she would die

[ananke-xiii]

what drives me a little insane about Charming Acres is that it's a male fantasy created by a scared white old man but it's also enforced by women (specifically the labor of women, more on Ms Dowling vs Cindy Smith later) who act a little bit like Mary/Eve, aka as regulators deciding who must stay (Cindy) and who must leave (Sunny). Much to think about that but also! It connects so wonderfully with my obession with the themes of "the danger/charm of families one marries into" and "access to the Normal World in SPN is always via a woman (and her labor)". The first pertains to John Winchester whose whole life was founded on lies upon lies, the second to his sons who seem to follow in his footsteps thinking that a woman would "fix" their issues and allow them to "escape" from their world, their reality. Women are fictions that can save (or damn), that can lock you out of heaven or welcome you back in. And, of course, these are all invented narratives (the madonna/whore dichotomy) but they also show how much these ideas are ingrained in the (male) characters' minds. They don't actually need a psychic making them "buy into" a fiction because, in a way, they've mind-controlled themselves into being slaves of these narratives . But they're also the perpetrators (the authors, if you will) of these narratives that they inflict upon others (for example, the women and men of Charming Acres) and, primarily, upon themselves (hence, Chip getting trapped in his own mind where he can be happy without harming others which, if we want to go there, can also translate into: in order to be happy you will inevitably hurt people). In other words, there is a "chip" inside these characters' minds that must be destroyed. So apart from madonna/whore (which is more related to possession and consent) the second interesting dichotomy in SPN is that of the master/slave (which is more related to mind control) which is, basically, the blueprint for all the other dichotomies in the show.

When, at the end of S14, Jack wills a world without lies into existence (committing yet another act of mind control) three things become violently apparent: the world itself is not as it is but it's as we narrate it to ourselves first and to others second, aka a fiction ("Breakdown" is also about this as well); lies are not only necessary but inherently essential for the world to function, aka lies are the basis of Order; Chuck doesn't simply act as a deus ex machina because, on the meta level, he's also an author here, an author that's actually needed in order to keep the illusion going, aka the world ceases to operate without a white man telling lies and pushing buttons. This means that there is no other option beyond the master/slave dichotomy which in turns means that the whole of S15 is just a display of suffering leading nowhere. Like, the show had made up its mind about the message it wanted to convey by "Moriah": you need an author because, without one, the world is in Chaos from which stories are impossible to tell because they need imagined and arbitrary beginnings and endings, aka they need some sort of Order AND "lies". Not only that, you need a hands-on God because without an author planning, scheming, deviating, inventing, interfering, aka being present, the story either ceases to exist or, if it still does exist, it's a boring story, a re-run nobody's really interested in anymore. And this is why, to me, all the meta madness of S15 is both cool (to a certain point) and dangerous (very much so) because it literalizes fiction which is always something very dangerous to do (see: the whole history of Christianity). But it also fictionalizes reality which is also very harmful, especially if the author god in question is a white man with close to limitless power and full-on unhingedness. So, in the end, women are fictions invented by men, about men and for other men, but so are men: fictions by men, about men and for other men. Ultimate reality is, therefore, the Fiction of Man, precisely of a white man, and there is no escape from that because the show cannot imagine power other than a relation between master and slave. And, look, my intentions are not to shit on SPN because at least a good 80% of TV shows are white man-centered and can't imagine power in any other way, but I hold a grudge towards its finale because I think SPN could've done a tiny bit better than describing Jack as "top dog" (which is in total and absolute contradiction with the fact that, apparently, Amara and Jack are in "harmony". I mean, come on, this is bad writing, period).

So Chuck's story, the Fiction of (White) Man can only "end" the same way it "began", which is to go back to the "In the beginning..." (and how many Genesis do we have in SPN?, it is a bit exhuasting, really, lol). In "Inherit the Earth" (and let's quickly remember who the subject of that sentence is: the meek, the meek shall inherit the earth), Chuck, not so subtly, reverted Earth to its "origins" and what we are left with are Dean, Sam and Jack, aka Adam, Eve and the snake. But this time there is no escape from this "Paradise" because, as I said above, the only ways to "escape" are via marrying into another family or via a woman's labor. In this episode, however, there is no "escape" because 1. the only family you can marry into is your own family (:|) and 2. there are no women (brrr). The only solution is, then, to go back to how things were before, to the status quo, by cheating and self-sacrifice.

Finally, in this light the "destiny/free will" dichotomy is reduced as follows: "destiny" is perceived as being (ab)used and controlled by a master (aka being a slave), therefore "free will" is being the one who (ab)uses and is in control (aka being the master). Which means that "free will" for one means "destiny" for another. Or: there is no real sense of collectiveness and community in the show (this is why the world must disappear and the only characters left are the ones mentioned above). I think, maybe, the answer is that, ultimately, control is never questioned because fundamentally the show thinks that control in the hands of one IS not only possible but necessary. And while it can go as far as to at least explore (to a very limited degree) why controlling someone else's body is (possession/consent etc), indeed, wrong, it just cannot do the same when it comes to a more "invisible" control, that of the mind. Which means that the show cannot imagine anything past the master/slave dichotomy. And I think this limitation was reinforced by the meta-fication of the show: on one hand it's a cool concept, on the other they put themselves into a corner because there is no show without someone actually controlling and performing the narrative both for an audience and for profit. Hence, the final shot where the fourth wall is broken, all masks are revealed and we're back to "Moriah" where Jack tells people to "stop lying": what remains is the audience feeling disappointed and realizing that we do want an Author because we want more fiction and we want our stories to make sense. We can even yield control for more fiction which is, of course, a human but also horrifying attitude (and by the way this is 100% relevant to contemporary discourses about generative AI and authorship so I predict SPN will keep interest people in the near future, mark my words).

Ultimately SPN feels "scammy" to me because it feels like Charming Acres: we as audience have been lured into this interesting fiction, the fiction we craved, but, at the very last second, the show has closed itself off into his own "fictionality". It's like "The Truman Show" where Truman is actually SPN that "excused" itself and went out the fake-sky door. Which turns us, the audience, into Christoff, the author/god/director who gets "defeated" by Truman leaving the stage, who gets to "spy" on the "author-less" characters against their fought for... "free will". SPN finale basically removes the author (here to be read as the people who create the story) as "filter" and leaves, on a meta level, the characters alone with the audience who, obvi, reveal themselves as actors because it's fiction, not reality, baby.

In other words, the audience can't "own" the story. We might be like Becky and, perhaps, even be better at writing that story but we're not its authors. Of course, this is totally an uncool way to end a story and a cruel low blow to the audience that actually made it even possible but, again, very relevant to the current scenario we live in: there is a metaphorical "chip" that feeds on our info, shapes our reality and is also generating our fictions. We're becoming authors, characters and audience of our own fictions but.... are they really our own? What do we see when we see our own narratives played out in front of our eyes? Can we claim authorship? Why do we want authorship and how fair is it to claim it as our own? Aren't we all a little bit characters trapped inside our minds? Can we escape from this? How does a manifest, individual "author" fare against an immanent, occult one? What are the alternatives we can come up with? Can we imagine power differently? These and many other questions may legitimately arise :P.

[inkintheinternet]

Ai Philosophers Thinkers of the New World

By Arjuwan Lakkdawala

Ink in the Internet

After writing had been established in the world - its origin going back to Mesopotamia (ancient Iraq) closely followed by Ancient Egypt and China, the records show Greek philosophers with overwhelming efforts trying to articulate into words mysteries of the natural world, reason and logic, concepts of what makes a good life, and of course questions about ethics and existence. In my research I found 9 Greek philosophers, to mention a few of them Pythagoras (notoriously famous for the Pythagoras Theorem).

Side Note: "Although the theorem has long been associated with Greek mathematician-philosopher Pythagoras (c. 570–500/490 BCE), it is actually far older. Four Babylonian tablets from circa 1900–1600 BCE indicate some knowledge of the theorem, with a very accurate calculation of the square root of 2."

Protagoras, Socrates, Plato, and Aristotle. It is said that western thought is highly influenced by them. As history unfolded we know that many philosophers expressed their ideas, and presently we have an infinite maze of ideologies.

In a free world everyone believes what they want to believe, but if it is harmful to society, causes self harm, is illogical, then it should be illegal and wrong.

Up to now we were doing all the thinking, but presently there is an emerging new class of very different thinkers, and that is Artificial Intelligence, computer scientists are taking big data which is basically infinite and feeding it to Ai, and concepts from neuroscience is being used to develope more powerful computing like neuromorphic chips.

A study published in Neuron in 2022 explains "Bioneural Circuitry." In simple words, neurons use electricity to send signals so it is possible to merge them with electrical signals from artificial intelligence on electrodes. This could be similar to the technology used by Neuralink implants.

A very futuristic idea that occurred to me is what if a hypothetical Ai (Star-Sci) is used as a teacher from kindergarten to university graduation, with constant upgrade of information as the years go by. This teacher would teach and even give tests. An Ai model that teaches everything. Students would instead of naming universities or a string of teachers, be known only as pupils of a certain Ai. This is of course a very sci-fi idea, but in realistic terms as I see anarchy increasing in the world, and the cases of teachers behaving inappropriately with students, I think Ai teachers will be looked upon favourably pretty soon.

After thought: academics can contribute for free or paid to the Star-Sci life long teaching system. This could be a new style university, where students only interect with Ai, preventing any non-educational use of the institute or corruption from fellow students or teachers.

We in this current age are going through the greatest social experiment. That is the Internet which is a merging of human mind with artificial intelligence. It is changing the world in ways we cannot predict as this hasn't happened before. We have no previous case studies.

Ai is reportedly often used in writing essays, I personally use search engines for my research. I avoid using Ai to do any of my writing, I don't want my mind to rely on artificial intelligence, because I believe it will corrode my much superior human intelligence.

I think artificial intelligence is good to use in the management of big data, but not in any paper with concepts within the ability of human calculation and comprehension.

Scientists are trying to come up with a unified theory of consciousness. According to research and most human experiences we have an outer sense in which we comprehend empirical evidance and what we immediately experience, and then we have an inner consciousness in which we think, form memories and plan ahead. The outer world is not really complicated, it's the inner world - thoughts that can warp into unknown manesfeststions, some of these neurological paths can lead to genuise and others to insanity.

This unknown manesfestations of consciousness is exactly what defines the difference between human thinking and Ai programming. Artificial intelligence cannot have thoughts at all. It can only compute programmed instructions based on the data it has. But with big data this programming could in the near future start to resemble consciousness. As we do not and mentally cannot compute big data, so the output of Ai from big data could seem to us like independent thoughts.

It is crucial to understand this about Ai as technology integrates it further in our lives and we interact with it increasingly.

At the same time we must avoid using it where we can use our human intelligence. This is of prime importance if we are to preserve the excellence of our cognitive abilities.

Neuroscience studies reveal that neural networks that are not used get lost. We don't really need scientific experiments to confirm this, as we can immediately tell when our memory improves, or we get better at any intellectual skill, well known that it only happens with practice, therefore if we rely on Ai it will result gradually in mental decline.

As we think our brains form complex neural networks that improve our intelligence regarding a study or concept.

Thoughts are different than feelings, we have often heard that the brain is logical and sensible while the heart isn't. Studies have confirmed that the heart has a neural network known as the intrinsic cardiac nervous system (ICNS). While this does not say that it is why we have feelings that could contradict our intelligence, it is an interesting hypothetical concept to explain how feelings and logical thoughts can be in conflict with each other.

Being irrational and illogical or hateful, and having several other destructive traits, that cause self harm, or harm to society, is unfortunately observed more often these days.

Mental health is a very complicated and difficult terrain to traverse, we are nothing without our consciousness. Our whole quality of life depends on the level of our intelligence and the health of our minds - this is maintained and achieved by a good upbringing, secure environment, healthy diet, and a rounded education with physical exercise.

There are foods that improve the health of our brains, and foods that increase the risk of Alzheimer's disease and dementia and have other negatives effects. Google for articles about brain health foods.

(I have added notes to this articles in the comments section).

Arjuwan Lakkdawala is an author and independent science researcher.

X/Twitter/Instagram: Spellrainia

Email: arjuwan.writer@gmail.com

Copyright ©️ Arjuwan Lakkdawala 2024

Sources:

MBF Bioscience - Researchers map and explore the heart's "little brain"

Psychology Today - Neuroplasticity - reviewed by Psychology Today staff

Medium - Neuromorphic Computing

ABclonal Knowledge Base - Melding Neuroscience with Computer Technology  - Kin Leung

Pythagorean Theorem - written and fact checked by the editors of Britannica Encyclopedia

Singularity Hub - Scientists are Working Towards a Unified Theory of Consciousness  - Shelly Fan

The Collector - 9 Greek Philosophers who Shaped the World - Eddie Hodsdon, BA professional Writing, member Canterbury Archealogical Trust

[jcmarchi]

DeepSeek's AI breakthrough: Fewer resources, big impact

New Post has been published on https://thedigitalinsider.com/deepseeks-ai-breakthrough-fewer-resources-big-impact/

DeepSeek's AI breakthrough: Fewer resources, big impact

On December 26th, a modest-sized Chinese company named DeepSeek introduced advanced AI technology, rivaling the top chatbot systems from giants like OpenAI and Google.

This achievement was noteworthy for its capability and the cost-efficiency with which it was developed. Unlike its large competitors, DeepSeek created its artificial intelligence, DeepSeek-V3, using significantly fewer specialized processors, which are typically essential for such advancements.

Cost efficiency and technological breakthrough

These processors are at the heart of a fierce tech rivalry between the U.S. and China. The U.S. aims to keep its lead in AI by restricting the export of high-end chips, such as those from Nvidia, to China.

However, DeepSeek’s success with fewer resources raises concerns about the effectiveness of U.S. trade policies, which have inadvertently spurred Chinese innovation using more accessible technologies.

DeepSeek-V3 impressively handles tasks like answering queries, solving puzzles, programming, and matching industry standards. Remarkably, it was developed with just around $6 million worth of computing resources, starkly contrasting the $100 million Meta reportedly invested in similar technologies.

Chris V. Nicholson from Page One Ventures pointed out that more companies could afford $6 million than the heftier sums, democratizing access to advanced AI technology.

Strategic implications and global impact of DeepSeek

Previously, experts believed only firms with substantial financial resources could compete with leading AI firms, which train their systems on supercomputers requiring thousands of chips.

DeepSeek, however, managed with just 2,000 chips from Nvidia. This efficient use of limited resources reflects the forced innovation resulting from chip restrictions in China, as Jeffrey Ding from George Washington University noted.

Recently, the U.S. tightened these restrictions to prevent China from acquiring advanced AI chips via third countries. This is part of ongoing efforts to limit Chinese firms’ potential military use of these technologies, which have resorted to stockpiling chips and sourcing them through underground markets.

ChatGPT vs Bard: What are the top key differences?

We’re taking a look at Bard vs ChatGPT and their key differences like technology, internet connection, and training data.

DeepSeek, a company rooted in quantitative stock trading, has been leveraging its profits to invest in Nvidia chips since 2021, fueling its AI research rather than consumer products. This focus has allowed it to bypass stringent Chinese regulations on consumer AI, attracting top talent and exploring diverse applications from poetry to complex examinations.

While leading U.S. firms continue to push AI boundaries, DeepSeek’s recent achievements underline its growing prowess in the field. It also highlights the broader shift towards open-source AI, gaining traction as companies like Meta openly share their technologies. This shift increasingly positions China as a central player in AI development, posing a strategic challenge to U.S. dominance in the field.

As the debate continues over the potential risks of open sourcing AI in the U.S., such as spreading misinformation, the global open source community, increasingly led by China, might shape the future of AI development, suggesting a significant geopolitical shift in the technology landscape.

Have you seen our 2025 event calendar?

From agentic AI to LLMOps, this year will be bigger than ever – join us in one of our 19 in-person events across the globe and network with other AI leaders.

Like what you see? Then check out tonnes more.

From exclusive content by industry experts and an ever-increasing bank of real world use cases, to 80+ deep-dive summit presentations, our membership plans are packed with awesome AI resources.

Subscribe now

[peterbordes]

Nvidia’s Jensen Huang says AI agents are ‘a multi-trillion-dollar opportunity’ at last nights CES AI keynote. Let that “multi Trillion” sink in, because this global movement is just getting started.

[memenewsdotcom]

Nvidia stock rises by record

[fastlivenostress]

Elon Musk's Ambitious Plan to Implant Brain Chips: What You Need to Know

Elon Musk, the visionary entrepreneur and founder of SpaceX, Tesla, and The Boring Company, has set his sights on an incredibly ambitious project: implanting brain chips into humans. This may sound like science fiction, but Musk’s Neuralink company has already made significant progress in developing a brain-computer interface that could revolutionize the way we interact with technology. In this…

[kingtainorman]

New Computer Breakthrough: Light-Speed Unlocked !

Photonics will take the world by storm in less than five years. That's my prediction. Why? Because AI is going to help build the platform that it will run on the most optimally. Also, it's already being implemented with big name companies.... It'll be mainstream after a few more innovations drive the use case scenarios up and the costs down.

[aitalksblog]

Top Weekly AI News – May 02, 2025

AI News Roundup – May 02, 2025 Nvidia says Anthropic is telling ‘tall tales’ in its defense of U.S. AI chip restrictions on China Nvidia and Anthropic are in a public disagreement over US AI chip export restrictions to China, with Anthropic supporting tighter controls while Nvidia criticizes the policy’s impact on competition cnbc May 01, 2025 Why the A.I. Race Could Be Upended by a Judge’s…

[walkingghotst]

Europe Artificial Intelligence (AI) Chip Market Size, Share, Comprehensive Analysis, Opportunity Assessment (2019-2027)

Europe Artificial Intelligence (AI) Chip Market is expected to grow from US$ 1.25 Bn in 2018 to US$ 16.04 Bn by the year 2027 with a CAGR of 33.0% from the year 2019 to 2027.

Europe Artificial Intelligence (AI) Chip Market Introduction

A key driver fueling the expansion of the European AI Chip market is the substantial capital investment directed towards artificial intelligence chip start-ups. The escalating demand for real-time consumer behavior insights, alongside the pursuit of enhanced operational efficiency, are also significant factors propelling the broader integration of AI across diverse industries. Furthermore, the anticipated incorporation of AI chips into edge computing devices is poised to further stimulate the market's growth throughout the forecast period. Across the globe, major industries spanning BFSI, retail, IT & telecom, automotive & transportation, healthcare, media & entertainment, manufacturing, government, and energy & power are actively embracing and investing in transformative technologies such as artificial intelligence, the Internet of Things (IoT), big data, and predictive analytics. This widespread adoption is a direct consequence of the demonstrated successes of AI applications, leading to improved operational efficiency, increased sales revenue, and enhanced interactions with customers.  

Download our Sample PDF Report

@ https://www.businessmarketinsights.com/sample/TIPRE00005730

Europe Artificial Intelligence (AI) Chip Strategic Insights

Strategic insights concerning the Europe Artificial Intelligence (AI) Chip market deliver a data-centric examination of the industry's structure, encompassing prevailing trends, key market participants, and specific regional characteristics. These insights offer practical recommendations, empowering readers to distinguish themselves from competitors by discovering unexploited market segments or formulating distinctive value propositions. By effectively utilizing data analytics, these insights assist industry stakeholders, whether they are investors, manufacturers, or other actors, in anticipating shifts in the market. A forward-looking viewpoint is crucial, aiding stakeholders in predicting market changes and strategically positioning themselves for sustained success in this evolving European region. Ultimately, impactful strategic insights equip readers to make well-informed decisions that foster profitability and support the realization of their business goals within the market.

Europe Artificial Intelligence (AI) Chip Regional Insights

The geographic scope of the Europe Artificial Intelligence (AI) Chip market defines the specific territories in which a business operates and competes. Comprehending local variations, such as diverse consumer preferences (for instance, the need for specific plug types or battery backup durations), differing economic landscapes, and regulatory frameworks, is vital for adapting strategies to particular markets. Businesses can broaden their market reach by identifying markets that are currently underserved or by modifying their offerings to align with local requirements. A focused market approach enables more effective resource management, targeted marketing efforts, and improved competitive positioning against local players, ultimately driving expansion in those targeted regions.

Europe Artificial Intelligence (AI) Chip Market Segmentation

Europe Artificial Intelligence (AI) Chip Market: By Segment

Data Center

Edge

Europe Artificial Intelligence (AI) Chip Market: By Type

CPU

GPU

ASIC

FPGA

Others

Europe Artificial Intelligence (AI) Chip Market: By Industry Vertical

BFSI

Retail

IT & Telecom

Automotive & Transportation

Healthcare

Media & Entertainment

Others

Europe Artificial Intelligence (AI) Chip Market: By Country

Germany

France

Italy

UK

Russia

Rest of Europe

Europe Artificial Intelligence (AI) Chip Market: Companies Mentioned

Advanced Micro Devices, Inc.

Alphabet Inc. (Google)

Huawei Technologies Co., Ltd.

IBM Corporation

Intel Corporation

Micron Technology, Inc.

NVIDIA Corporation

Qualcomm Incorporated

Samsung Electronics Co., Ltd.

Xilinx, Inc.

About Us:

Business Market Insights is a market research platform that provides subscription service for industry and company reports. Our research team has extensive professional expertise in domains such as Electronics & Semiconductor; Aerospace & Defense; Automotive & Transportation; Energy & Power; Healthcare; Manufacturing & Construction; Food & Beverages; Chemicals & Materials; and Technology, Media, & Telecommunications

[asharid]

Chip Kuantum: Masa Depan Teknologi Komputasi oleh Microsoft

Dunia teknologi terus berkembang dengan cepat, dan salah satu inovasi paling menarik yang muncul adalah komputasi kuantum. Microsoft telah mengambil langkah besar dalam bidang ini dengan mengembangkan chip kuantum, sebuah teknologi revolusioner yang berpotensi mengubah cara kita memproses data. Artikel ini akan membahas konsep chip kuantum, keunggulan teknologi kuantum dibandingkan komputasi…

[youthchronical]

Intel Names Lip-Bu Tan as New CEO Amid Turnaround Efforts

Intel, a fallen Silicon Valley icon trying to restore its reputation as America’s most prominent semiconductor company, has named Lip-Bu Tan, an experienced business and technology leader, as its new chief executive. Mr. Tan, 65, will be responsible for reviving the fortunes of a chip-making company that has fallen from grace. Once one of the best-known names in technology, the semiconductor…

[jcmarchi]

MIT engineers grow “high-rise” 3D chips

New Post has been published on https://thedigitalinsider.com/mit-engineers-grow-high-rise-3d-chips/

MIT engineers grow “high-rise” 3D chips

The electronics industry is approaching a limit to the number of transistors that can be packed onto the surface of a computer chip. So, chip manufacturers are looking to build up rather than out.

Instead of squeezing ever-smaller transistors onto a single surface, the industry is aiming to stack multiple surfaces of transistors and semiconducting elements — akin to turning a ranch house into a high-rise. Such multilayered chips could handle exponentially more data and carry out many more complex functions than today’s electronics.

A significant hurdle, however, is the platform on which chips are built. Today, bulky silicon wafers serve as the main scaffold on which high-quality, single-crystalline semiconducting elements are grown. Any stackable chip would have to include thick silicon “flooring” as part of each layer, slowing down any communication between functional semiconducting layers.

Now, MIT engineers have found a way around this hurdle, with a multilayered chip design that doesn’t require any silicon wafer substrates and works at temperatures low enough to preserve the underlying layer’s circuitry.

In a study appearing today in the journal Nature, the team reports using the new method to fabricate a multilayered chip with alternating layers of high-quality semiconducting material grown directly on top of each other.

The method enables engineers to build high-performance transistors and memory and logic elements on any random crystalline surface — not just on the bulky crystal scaffold of silicon wafers. Without these thick silicon substrates, multiple semiconducting layers can be in more direct contact, leading to better and faster communication and computation between layers, the researchers say.

The researchers envision that the method could be used to build AI hardware, in the form of stacked chips for laptops or wearable devices, that would be as fast and powerful as today’s supercomputers and could store huge amounts of data on par with physical data centers.

“This breakthrough opens up enormous potential for the semiconductor industry, allowing chips to be stacked without traditional limitations,” says study author Jeehwan Kim, associate professor of mechanical engineering at MIT. “This could lead to orders-of-magnitude improvements in computing power for applications in AI, logic, and memory.”

The study’s MIT co-authors include first author Ki Seok Kim, Seunghwan Seo, Doyoon Lee, Jung-El Ryu, Jekyung Kim, Jun Min Suh, June-chul Shin, Min-Kyu Song, Jin Feng, and Sangho Lee, along with collaborators from Samsung Advanced Institute of Technology, Sungkyunkwan University in South Korea, and the University of Texas at Dallas.

Seed pockets

In 2023, Kim’s group reported that they developed a method to grow high-quality semiconducting materials on amorphous surfaces, similar to the diverse topography of semiconducting circuitry on finished chips. The material that they grew was a type of 2D material known as transition-metal dichalcogenides, or TMDs, considered a promising successor to silicon for fabricating smaller, high-performance transistors. Such 2D materials can maintain their semiconducting properties even at scales as small as a single atom, whereas silicon’s performance sharply degrades.

In their previous work, the team grew TMDs on silicon wafers with amorphous coatings, as well as over existing TMDs. To encourage atoms to arrange themselves into high-quality single-crystalline form, rather than in random, polycrystalline disorder, Kim and his colleagues first covered a silicon wafer in a very thin film, or “mask” of silicon dioxide, which they patterned with tiny openings, or pockets. They then flowed a gas of atoms over the mask and found that atoms settled into the pockets as “seeds.” The pockets confined the seeds to grow in regular, single-crystalline patterns.

But at the time, the method only worked at around 900 degrees Celsius.

“You have to grow this single-crystalline material below 400 Celsius, otherwise the underlying circuitry is completely cooked and ruined,” Kim says. “So, our homework was, we had to do a similar technique at temperatures lower than 400 Celsius. If we could do that, the impact would be substantial.”

Building up

In their new work, Kim and his colleagues looked to fine-tune their method in order to grow single-crystalline 2D materials at temperatures low enough to preserve any underlying circuitry. They found a surprisingly simple solution in metallurgy — the science and craft of metal production. When metallurgists pour molten metal into a mold, the liquid slowly “nucleates,” or forms grains that grow and merge into a regularly patterned crystal that hardens into solid form. Metallurgists have found that this nucleation occurs most readily at the edges of a mold into which liquid metal is poured.

“It’s known that nucleating at the edges requires less energy — and heat,” Kim says. “So we borrowed this concept from metallurgy to utilize for future AI hardware.”

The team looked to grow single-crystalline TMDs on a silicon wafer that already has been fabricated with transistor circuitry. They first covered the circuitry with a mask of silicon dioxide, just as in their previous work. They then deposited “seeds” of TMD at the edges of each of the mask’s pockets and found that these edge seeds grew into single-crystalline material at temperatures as low as 380 degrees Celsius, compared to seeds that started growing in the center, away from the edges of each pocket, which required higher temperatures to form single-crystalline material.

Going a step further, the researchers used the new method to fabricate a multilayered chip with alternating layers of two different TMDs — molybdenum disulfide, a promising material candidate for fabricating n-type transistors; and tungsten diselenide, a material that has potential for being made into p-type transistors. Both p- and n-type transistors are the electronic building blocks for carrying out any logic operation. The team was able to grow both materials in single-crystalline form, directly on top of each other, without requiring any intermediate silicon wafers. Kim says the method will effectively double the density of a chip’s semiconducting elements, and particularly, metal-oxide semiconductor (CMOS), which is a basic building block of a modern logic circuitry.

“A product realized by our technique is not only a 3D logic chip but also 3D memory and their combinations,” Kim says. “With our growth-based monolithic 3D method, you could grow tens to hundreds of logic and memory layers, right on top of each other, and they would be able to communicate very well.”

“Conventional 3D chips have been fabricated with silicon wafers in-between, by drilling holes through the wafer — a process which limits the number of stacked layers, vertical alignment resolution, and yields,” first author Kiseok Kim adds. “Our growth-based method addresses all of those issues at once.” 

To commercialize their stackable chip design further, Kim has recently spun off a company, FS2 (Future Semiconductor 2D materials).

“We so far show a concept at a small-scale device arrays,” he says. “The next step is scaling up to show professional AI chip operation.”

This research is supported, in part, by Samsung Advanced Institute of Technology and the U.S. Air Force Office of Scientific Research.