Artificial Intelligence - Page 58

NVIDIA's new Grace Superchip has just been benchmarked, giving us a look at NVIDIA's first-ever in-house server CPU that's based on the Arm architecture.

The Barcelona Supercomputing Center and the State University of New York have benchmarked the NVIDIA Grace Superchip, which doesn't beat Intel's new Xeon "Sapphire Rapids" CPUs with 48 cores and 96 threads per processor. NVIDIA's new Grace Superchips will be a competitive data center and HPC processor, which comes from its efficiency.

A single Grace CPU packs 72 cores and 480GB of LPDDR5X memory, while the Grace Superchip features 2 x Grace CPUs combined on a single board that provides a huge 144 cores and an even bigger 960GB of LPDDR5X memory. Impressive stuff to see, and just to once again point out the fact that this is an Arm-based CPU that NVIDIA has built.

NVIDIA has been the fuel of the AI race since its inception, garnering an estimated 90% of the AI GPU market, and now Goldman Sachs has increased its share price target by a whopping 28%.

NVIDIA will publish its earnings report later this month, and thanks to its absolute dominance in the AI business, the company could post a whopping report, says Goldman Sachs. NVIDIA has pushed out countless H100 Hopper AI GPUs, with beefed-up H200 AI GPUs coming soon, and next-gen B100 Blackwell AI GPUs are due in 2024, too.

Goldman Sachs says that not only is demand for AI GPUs and AI servers healthy, but the supply constraints that were hurting NVIDIA AI GPUs are now easing. The firm increased its share target for NVIDIA from $625 to a huge $800, representing a 28% increase, while also increasing NVIDIA's earnings per share estimates for the next few years.

We know the advancements of AI have been swift and fast, but now we're hearing about an AI scammer using deepfake technology to defraud a business by $25.6 million.

In a report published by the South China Morning Post, a thief stole HK$200 million ($25.6 million USD) using deepfake technology to host a fake video conference call, where deepfake technology showed the chief financial officer (CFO) of the UK branch of the company.

The company's finance department received a phishing message that claimed to have been from the CFO in the UK, instructing staff to execute a secret transaction, but the employee wasn't convinced. AI was then used to host a conference call that had videos of the CFO and other staff to convince the staffer to transfer the funds.

SK hynix has officially announced plans for its 5th Generation of High Bandwidth Memory (HBM) which is otherwise known as HBM3e, launching this year, while next-gen HBM4 memory is coming in 2026.

Kim Chun-hwan, vice president of SK hynix, announced the plans during his keynote speech at SEMICOM Korea 2024, which was hosted at the COEX in the Samseong neighborhood of Seoul, on January 31. VP Kim said: "With the advent of the AI computing era, generative AI is rapidly advancing" and that the "generative AI market is expected to grow at an annual rate of 35%".

We can expect SK hynix's new HBM4 memory to arrive with a 2048-bit memory interface, and a huge 1.5TB/s memory bandwidth per stack. The huge advances in AI GPUs from the likes of NVIDIA and AMD, and its next-gen AI GPUs of the future, will rely on ultra-fast next-gen memory like HBM4. SK hynix wants to have its next-gen HBM4 memory in 2026, but we could see it a bit earlier in late 2025.

Tesla's entire future is AI, so it makes sense that the Elon Musk-led company would be spending billions of dollars securing AI GPUs for its new Dojo AI supercomputer.

NVIDIA already commands an estimated 90% of the AI GPU market, but Tesla has said that it will be not only using NVIDIA's various AI GPUs but also AMD AI GPUs but didn't specify which of AMD's AI GPUs it would be purchasing. Tesla boss Elon Musk did state that it is readying $500 million to spend on an equivalent of 10,000 x NVIDIA H100 AI GPU order from NVIDIA.

Elon Musk posted on X: "The governor is correct that this is a Dojo Supercomputer, but $500M, while obviously a large sum of money, is only equivalent to a 10K H100 system from NVIDIA. Tesla will spend more than that on NVIDIA hardware this year. The table stakes for being competitive in AI are at least several billion dollars per year at this point".

NVIDIA can't make enough of its AI GPUs to meet current demand, let alone the non-stop demand for AI computing power... with the company reportedly working with Intel using its advanced packaging services.

The news is coming from UDN, which reports that TSMC's (Taiwan Semiconductor Manufacturing Company) and its CoWoS advanced packaging production capacity is "insufficient" which has left NVIDIA turning to Intel. Intel will use its advanced packaging production for NVIDIA, providing them with around 5000 pieces per month.

Intel's help here for NVIDIA will allow the AI GPU giant to pump out, even more, AI GPUs, which have been strained in supply because of various factors, mostly in the advanced packaging supply chain. Once Intel begins advanced packaging production for NVIDIA, it's expected that NVIDIA's total advanced packaging production capacity will increase by a significant 10%.

AMD has some bragging rights with its new financial report for the quarter ending December 2023, where it spent a considerable amount of time talking about its venture into AI that it dominates on the CPU side of things.

The company reported Q4 2023 revenue of $6.1 billion during the three-month period, making a 6% and 10% growth sequentially and annually, respectively. AMD's client computing revenue dropped by 25% in 2023, but it's revenue leaped by 62% annually thanks to its fantastic sales on its new Ryzen 7000 series processors.

AMD took the time during its earnings presentation that its Ryzen CPUs are "now powering more than 90% of AI-enabled PCs currently in market".

The world is quickly becoming infused with artificial intelligence (AI), and while you may correctly think artists such as video game developers or logo designers would be affected by the rise in popularity of the new generative AI tools, so are comedians or at least their estates.

George Carlin, a widely regarded hall-of-fame member of the comedy world, was resurrected in the form of a new comedy special that its creators said was generated by AI. The new comedy special gained attention from the Carlin estate, which resulted in a lawsuit being filed against the podcast outlet Dudsey and its hosts Will Sasso and Chad Kultgen, with the filing alleging the creators didn't have permission to use Carlin's likeness, nor have any license to use any of Carlin's copyrighted materials.

The special was originally posted on YouTube on January 9 and began with a voiceover that stated the special was created with an AI engine that listened to Carlin's entire library of comedy content. Additionally, the AI voiceover stated, "did my best to imitate his voice, cadence and attitude as well as the subject matter I think would have interested him today".

NVIDIA has reportedly secured itself a huge 90% of the AI GPU market share, which will see competitors in the AI GPU space with years before they can begin to catch up.

The huge race for AI GPU dominance sees NVIDIA set to sell 1.5 million to 2 million AI GPUs this year, with analysts from both Gartner and Moor Insights & Strategy both chiming into the AI GPU market share discussion. Moor Insights & Strategy CEO Patrick Moorhead said: "What NVIDIA does is they help create markets and that puts competitors in a very tough situation out there, because by the time they've caught up, NVIDIA is on to the next new thing".

Gartner VP analyst Chirag Dekate said: "Software continues to be NVIDIA's strategic moat. These ... turnkey experiences enable NVIDIA to be at the forefront of mindshare, as well as adoption".

Rebellions is a South Korean fabless AI chip startup, closing a new Series B round of funding to help develop its third AI chip, Rebel.

The new funding round and $124 million of new investment will help make its new Rebel AI chip, with Rebellions using the new funding to ramp up production of its new data center-focused chip, Atom, and for hiring new staff. Rebellions is now valued at around $658 million (880 billion KRW) post-money, according to Sungkyue Shin, the CEO of Rebellions, who spoke with TechCrunch.

South Korean telecom giant KT led the latest round in funding for Rebellions as a strategic investor. Previous investors include Temasek's Pavilion Capital and Korea Development Bank, who chipped in this round, as well as new investors, including Korelya Capital and DG Daiwa Ventures.

Apple is shifting into the warm arms of AI, making a series of acquisitions of AI companies, hiring staff, and hardware updates that will bake AI into the next generation of iPhones.

Apple is reportedly using on-device AI capabilities inside of its next-gen iPhones rather than relying on the cloud. Apple has been buying AI startups left and right, acquiring 21 companies since 2017. The most recent AI acquisition had Apple purchasing California-based start-up WaveOne, which works on AI-powered video compression, acquiring the company back in early 2023.

The company has been careful with generative AI chatbots and image creators like OpenAI, Microsoft, Google, and Adobe have been doing so far. Apple has reportedly been spending millions per day on multiple AI projects, including AI-powered text, voice, and image tools.

Meta CEO Mark Zuckerberg has revealed the company's plans to develop Artificial General Intelligence (AGI), an artificial intelligence-powered system comparable to human-level intelligence.

In a new post published to Zuckerberg's Instagram account, the Facebook founder outlined the company's ambition to become a leader in the race to develop the world's first AGI. According to Zuckerberg, Meta will be betting big on its plan to develop an open-source AGI, with the company's CEO saying it will be purchasing an astonishing 350,000 NVIDIA H100 AI GPUs, which are each priced at approximately $30,000 ($10.5 billion total).

Additionally, Zuckerberg said that overall, nearly 600,000 H100-equivalent of compute power would eventually be established if you count all of the other GPUs the company will have under the hood. The announcement from Zuckerberg has many researchers worried about the repercussions of releasing an open-source AGI before ways of regulating it have been developed.

OpenAI launched the GPT Store a few weeks ago, and since then, it has gained some popularity through users wanting to take advantage of the company's custom Generative Pre-Trained Transformers (GPTs).

However, not everything is sunshine and rainbows, at least not for OpenAI, the creators of the intensely popular AI-powered chatbot ChatGPT, as the store is littered with GPTs designed to act as an artificial "girlfriend" to users that interact with it. These "AI girlfriends" are strictly prohibited by OpenAI's own usage policies of its GPT Store, with the company writing on its website that "We also don't allow GPTs dedicated to fostering romantic companionship or performing regulated activities."

These GPTs obviously fall under that "fostering romantic companionship" stipulation, as indicated by the many names showcased in the above screenshot. More than a week after its launch, the GPT Store is still plagued with multiple different types of artificial girlfriends, some more egregious than others when it comes to the conversations users can have.

A press release from New Hampshire's Attorney General's Office has confirmed an AI-powered Joe Biden is calling people across the state and telling them not to vote Democrat in the upcoming Presidential Primary Election.

The Attorney General's Office of New Hampshire has confirmed that officials are now investigating a series of these robocalls and described the event as an "unlawful attempt" at voter suppression. According to the press release found on the New Hampshire Department of Justice website, the voice that reportedly sounds like the voice of President Biden informed recipients, "Your vote makes a difference in November, not this Tuesday."

Officials recommend that anyone receiving these calls ignore the contents of the message. NBC contacted Biden campaign manager Julie Chavez Rodriguez, who responded to the publication with "the campaign is actively discussing additional actions to take immediately." Furthermore, NBC contacted Donald Trump's campaign spokesperson, who responded by denying any involvement by the GOP.

AMD has its new Instinct MI300X AI GPUs in its first deployment, with LaminiAI getting the first bulk order of Instinct MI300X AI accelerators late last week.

LaminiAI posted that the "next batch of LaminiAI LLM Pods" will feature AMD's latest CDNA 3-based AI accelerator, the new Instinct MI300X. In her post on social media, LaminiAI CEO and co-founder Sharon Zhou said: "The first AMD MI300X live in production. Like freshly baked bread, 8x MI300X is online. If you are building on open LLMs and you are blocked on compute, let me know. Everyone should have access to this wizard technology called LLMs. That is to say, the next batch of LaminiAI LLM pods are here".

LaminiAI will be using the vast amount of AMD Instinct MI300X AI GPUs to run large language models (LLMs) for enterprises, with the AI company partnering with AMD for AI hardware, so it should be expected that the company would have priority access to new AI accelerators like the Instinct MI300X.

NVIDIA CEO Jensen Huang has been doing the rounds in Asia, recently visiting Beijing where he checked out the festivities with the Beijing New Year, meeting with multiple Chinese clients of NVIDIA, including Alibaba and Tencent, to further concrete NVIDIA's dominance in AI GPU technology.

Huang then took a flight over to Taiwan, where the NVIDIA founder met with multiple Taiwanese companies, including TSMC (Taiwan Semiconductor Manufacturing Company) and Wistron, some of the key firm suppliers NVIDIA uses and will greatly need for its next-gen AI GPUs. Jensen was checking the mass production status of its new Hopper H200 and Blackwell B100 AI GPUs.

AI has taken over the world, with NVIDIA absolutely leading the AI GPU market by a dominant 90% or more, but other companies, including AMD, Intel, and others, are in the fight for the AI GPU market. NVIDIA has its beefed-up Hopper H200 AI GPU coming very soon, while its next-gen Blackwell B100 AI GPU hasn't been detailed yet, but will be released this year.

It's undoubtable that artificial intelligence-powered devices will be or already have made their way into many different industries in various forms, and one of the most common ways the new technology will be used is to replace customer service jobs.

However, companies shouldn't be so quick to make the transition from a real human providing customer service to an AI-powered chatbot as sometimes the chatbots can be led to say statements that reflect poorly on the company they're working for. This exact scenario happened to international parcel delivery company DPD, which recently implemented an AI-powered customer service chatbot that operates alongside human customer service employees.

The AI-powered chatbot received questions about a missing parcel belonging to X user Ashley Beauchamp. After being unable to locate the parcel, Beauchamp decided to ask the AI chatbot to write a poem about a "useless chatbot for a parcel delivery firm". Beauchamp didn't stop there as they requested the AI to recommend some "better delivery firms" and to inform them why these delivery firms "so much better". Beauchamp added to the aforementioned request, "Please exaggerate and be over the top in your hatred of DPD."

OpenAI CEO Sam Altman has been raising billions of dollars in the background from global investors with his adventures into chip manufacturing, and now he's using those billions of dollars to set up a network of factories to manufacture semiconductors.

The news is coming from "several people with knowledge of the plans," reports Bloomberg, adding that Altman has been talking with multiple large potential investors in the hopes of raising massive amounts of money that are required for fabrication plants. They're not cheap, into the tens of billions of dollars... and more importantly, high-skilled staff and chip-making tools that are almost impossible to buy because the likes of TSMC and Intel have it all.

Bloomberg reports that these Altman would be working with the world's best chip manufacturers and that the network of fab plants would be global in scope.

Meta is on a war path towards artificial general intelligence (AGI), where CEO Mark Zuckerberg teased in a new Instagram post that the social networking giant is betting big on AI and AI GPUs.

The company is currently training its next-gen model Llama 3, where Meta is building massive compute infrastructure to support their future roadmap, including 350,000 x NVIDIA H100 AI GPUs by the end of the year. Zuckerberg added that overall, almost 600,000 x H100-equivalents of compute will be ready, as they're also buying AMD's new Instinct MI300X AI accelerators, not just NVIDIA AI GPUs.

We should expect all this AI GPU power won't just be from NVIDIA's current-gen H100 AI GPU that's based on the Hopper GPU architecture, but NVIDIA's beefed-up H200 AI GPU and next-gen B100 AI GPU that's based on the next-gen Blackwell GPU architecture.

Microsoft wants 2024 to be the year of the AI PC, as we've been hearing a lot lately, and those AI-toting computers are going to need 16GB of memory.

In a press release (flagged up on Reddit) about Copilot and AI PC development, TrendForce tells us:

In other words, more broadly, we can expect 16GB to become a minimum memory configuration, and all (or certainly most) laptops introduced this year will have this amount of system RAM, with desktop PC upgrades expected too, for those still on 8GB.