Artificial Intelligence - Page 41

A new report from Proof News alleged Apple, NVIDIA and other big tech companies used a dataset that contained copyrighted IP to train their respective AI models. That copyrighted IP included transcripts of YouTube videos from prominent creators, such as MKBHD, one of the platforms biggest technology reviewers.

The report cited an investigation into the dataset known as Pile, with the reporters claiming to have discovered transcripts or subtitles of more than 170,000 YouTube videos across 40,000 different channels. Some of those videos were from creators such as MrBeast, MKBHD, Jimmy Kimmel, Stephen Colbert, PewDiePie and many others. The report also revealed statements from companies stating they used the Pile dataset in the training of their AI models, as the dataset is free and open for public use.

This newly surfaced report raises the question of what happens to AI companies that use datasets containing copyrighted IP to train their AI models. Is the owner of the AI model responsible or the company that formed the dataset? Or both? OpenAI was caught in the hot water bath that is AI models and copyrighted data only a few months ago when Chief Technology Officer (CTO) Mira Murati was unable to answer whether OpenAI uses YouTube videos to train its AI models.

President Trump and his allies are working on a sweeping AI executive order that would see a bunch of "Manhattan Projects" to develop next-generation military technology, and "Make America First in AI".

A second Trump administration would not just kick off the series of Manhattan Projects developing next-gen military technology, but immediately review "unnecessary and burdensome regulations", that would see President Trump looking at AI policies that would help Silicon Valley companies and investors, reports the Washington Post.

The new framework would also create new "industry-led" agencies that would look at AI models and secure systems from foreign countries, according to a "copy of the document viewed exclusively by The Washington Post". The framework itself includes a section that's titled "Make America First in AI" which has a "markedly different strategy for the booming sector than that of the Biden administration, which last year issued a sweeping executive order that leverages emergency powers to subject the next generation of AI systems to safety testing".

Samsung's new HBM3E memory has reportedly passed NVIDIA certification, with new Samsung HBM3E memory supply expected to flow in the coming months.

A new report from UDN reveals that up to 30% of existing DRAM production capacity will be allocated to provide HBM3E memory, which is causing a huge effect of capacity crowding, which UDN reports "hurry up stocking (DRAM)" is expected to trigger a DRAM price increase wave.

Taiwanese memory manufacturers like Nanya, ADATA, and TEAMGROUP are "enjoying the benefits of price increases". Samsung is the world's memory leader, with a DRAM market share that's reportedly over 45%, with the South Korean giant to allocate 30% of its production capacity to make HBM3E memory for AI GPUs, with over 13% of the world's existing DRAM production capacity no longer invested into DDR4 and DDR5 and other DRAMs, which will reportedly lead to a decline in the DRAM market, with supply tighter than ever before.

AMD and Intel are about to unleash next-generation CPUs and APUs with beefed-up AI capabilities, but do you really need, or even want them? Probably not, and that's exactly how 84% of users voted in a new poll.

TechPowerUp posted a new poll asking if PC users were interested in paying for AI-enhanced hardware, but 84% of the votes said NO. Over 26,000 people voted, with 7% saying that they would and 9% not sure what they wanted to do.

W1zzard posted: "AI capabilities are becoming increasingly integrated into hardware devices, promising enhanced performance and functionality. However, this advanced technology often comes at a premium price. Would you pay more for hardware with AI features?"

We shouldn't really be shocked by the outcomes of surveys at this point, but we were rather taken aback by the apparent level of misconception out there regarding AIs like ChatGPT - with many folks seemingly believing that these chatbots have "some degree of consciousness."

According to a study from the University of Waterloo (flagged by TechSpot), two-thirds of respondents in a survey (of 300) in the US felt this was the case, and further they agreed that such AI tools can have "subjective experiences such as feelings and memories."

Of course, these are Large Language Models (LLMs) and they most certainly don't experience feelings - not by any definition or philosophy we're aware of - but they are cleverly constructed AIs that can appear this way, sure. Plus the datasets they're trained on are inevitably human content - the opinions and thoughts that they hoover up by the ton, from every corner of the web - so that's reflected in the replies to queries, clearly.

In early April, YouTube sent a clear message to AI model developers that downloading data from the platform and using it to train AI models is a clear violation of YouTube's terms of service.

This sentiment was reinforced in the same week as YouTube's public comment about its content being used to train AI model, but it came from a Google spokesperson who told the New York Times any, "unauthorized scraping or downloading of YouTube content" is prohibited. However, a new report from Proof News has found YouTube has been scraped for its data, and some of the biggest tech companies advancing AI have used it to train models.

According to a Proof News investigation, subtitles from 172,535 YouTube videos were siphoned from more than 48,000 channels, and some of these channels included prominent creators on the platform such as MKBHD (19 million subscribers), MrBeast (289 million), Jacksepticeye (31 million), PewDiePie (111 million), Stephen Colbert, John Oliver, Jimmy Kimmel, and more. Notably, the video transcriptions are subtitles files.

Microsoft has struck a $1.5 billion deal with Group 42 (G42), an artificial intelligence research and development firm operating in the United Arab Emirates, and is chaired by national security advisor His Highness Sheikh Tahnoun bin Zayed Ai Nahyan.

The deal has sparked security concerns, with two House committee chairs now sending a public letter to the White House prompting it to investigate the deal, following Sheikh Mohamed bin Zayed Al Nahyan's visit to Beijing to strengthen AI-related ties, and G42 being investigated less than a year ago for its association with China. The $1.5 billion deal has caused fear among US intelligence officials as it could mean advanced US AI technologies could eventually reach China.

Notably, UAE's AI minister admitted the US concerns about the deal are valid, as it would be for "any country that has adversaries". Despite these seemingly widespread concerns, Microsoft and G42 are making an effort to implement as much transparency about the deal as they can, with Microsoft President Brad Smith saying G42 won't be gaining any AI access to proprietary US technologies such as processors, AI model design tools, and more. Moreover, the Microsoft president said the UAE's access would be in a "vault within a vault".

Samsung will use its in-house 4nm foundry process to mass produce its next-generation HBM4 memory to directly take on South Korean competitor SK hynix and TSMC in the race for AI memory supremacy.

In a new report from the Korea Economic Daily, we're hearing that Samsung will use its 4nm foundry process for the logic die of HBM4 memory chips (sixth-generation HBM) according to industry sources of KED. The logic die itself sits at the base of the stacks of dies and is one of the core components of an HBM chip used on AI chips.

SK hynix, Samsung, and Micron all make HBM, with logic dies used on the latest HBM3E memory, but the new HBM4 memory of the future requires a foundry process that's ready with customized functions required by AI chip makers like NVIDIA, AMD, and Intel, and more.

Intel has reportedly placed 3nm orders with TSMC for its next-generation AI chip codenamed "Falcon Shores" which teases TSMC's new CoWoS advanced packaging will be used, with Falcon Shores expected to be produced in late 2025.

In a new report from CNyes, TSMC has received yet another "successful order" for Intel's next-gen AI chip called Falcon Shores on TSMC's newer 3nm process node and CoWoS advanced packaging technology, with Falcon Shores to take on NVIDIA's tight grip on the AI chip market.

The new Falcon Shores AI chip design has been finalized (tape out) and will reportedly enter mass production at the end of 2025. Intel acquired Habana back in 2019, with Habana maintaining its independence in its operating model, but over the years, things have changed and now Intel will combine Habana's technology with its own GPU technology.

SK hynix is gearing up to strengthen its "triangular alliance" with NVIDIA and TSMC for the future of AI GPUs and future-gen HBM4 memory.

The South Korean company will attend SEMICON Taiwan on September 4, with SK hynix president Kim Joo-sun to deliver a keynote speech at the CEO Summit of SEMICON Taiwan, which will be the first time that the company has taken a large role like this at the event.

Around 1,000 companies attend SEMICON, a major event in the semiconductor space. Companies like TSMC attend to show off their latest semiconductor equipment and technologies, build new relationships, and strengthen old ones.

MediaTek is plotting along with new server processors based on the Arm architecture, in both CPU and GPU form, using TSMC's newer 3nm process node. The new Arm-based MediaTek AI chips will be launched in the second half of 2025.

In a new report from UDN, we hear that mass production is aiming for the second half of 2025, with orders expected from large cloud service providers (CSPs). MediaTek hasn't responded to the rumors, of course, but industry analysis shows that the AI server market is "rising rapidly," reports UDN, and that high-end AI models require high-performance computing (HPC) chips from major manufacturers like NVIDIA and AMD.

However, HPC consumes far too much power and doesn't require a large amount of AI inference, so in this field, there's no need for HPC chips. The mid-to-low-end AI server market is now growing, generating new demand. The low-power Arm architecture processor has become a "new target" for major CSP manufacturers.

TSMC is reportedly preparing to start production of NVIDIA's next-gen Blackwell AI GPU platform, with strong customer demand, NVIDIA has reportedly increased AI chip investment in TSMC by 25%.

In a new report from UDN, we hear that NVIDIA has amped up Blackwell AI GPU orders by 25% at TSMC. This shows that the insatiable AI demand isn't slowing down. TSMC's performance in the second half of 2024 is going to be bonkers, and 2025 will be even bigger (for both companies).

NVIDIA's next-generation Blackwell AI GPU family will usher in new performance levels, with major manufacturers like Amazon, Dell, Google, Meta, Microsoft, and more to use Blackwell AI GPUs in their new AI servers, and right now the capacity exceeds expectations.

Samsung Foundry has proudly announced that it has secured its first 2nm AI chip order from a Japanese company, taking one of TSMC's long-standing customers away.

In a new press release, the South Korean electronics giant said it secured a Japanese company to fab its new 2nm AI chip, with Taejoong Song, Corporate VP at Samsung Electronics explaining: "This order is pivotal as it validates Samsung's 2nm GAA process technology and Advanced Package technology as an ideal solution for next-generation AI accelerators. We are committed to closely collaborating with our customers ensuring that the high performance and low power characteristics of our products are fully realized".

The Japanese company in question is Preferred Networks, a leading Japanese AI company that is involved the R&D focused on deep learning workloads. Preferred Networks is heading towards vertically integrating "the AI value chain from chips to supercomputers" with the Japanese company providing a medium for businesses to have their own in-house AI clusters.

Amazon AWS has just shown off its fourth-generation general purpose processor -- the new Graviton4 -- with 3x the compute power and memory over Graviton3.

The next-gen Graviton4 chip also has 75% more memory bandwidth than Graviton3 and 30% more performance. Amazon AWS adds that the new Graviton4 processors are rented out for $0.02845 per second of compute power. Yahoo! Finance reports that the price-performance ratio is "crucial" for AWS, as it uses its proprietary chips to power its cloud infrastructure and servers.

Rahul Kulkarni, Amazon's director of product management for Compute and AI, said: "Collectively it's delivering more price performance, which means for every dollar spent, you get a lot more performance".

The race to create the world's most powerful artificial intelligence system has certainly heated up, and by the looks of things, it isn't stopping anytime soon, with multiple tech companies flocking to NVIDIA for its powerful workstation GPUs used to train the AI models.

It was only a few months ago Musk confirmed it was building a massive AI factory with Dell and Supermico, and within these huge server farms will be NVIDIA GPUs that will be used to train xAI's model Grok. To "catch up" to the competition and make Grok a viable AI solution that is as good, if not better than leading models from OpenAI and Microsoft, Musk plans on throwing 100,000 Hopper-based GPUs into a server.

The SpaceX founder explained via a post on X that xAI contracted 24,000 H100 (Hopper) GPUs from Oracle, which are being used to train Grok 2. However, Musk said xAI will move forward with its 100,000 H100 system by itself as that will result in the "fastest time to completion". Moreover, Musk wrote the decision to go ahead with the 100,000 GPU system internally was the company's "fundamental competitiveness depends on being faster than any other AI company."

What is just a portion of the undiscovered bot farms plaguing the internet has been discovered by the Department of Justice (DOJ) and seized.

The DOJ announced on Tuesday that it seized a bot farm linked to Russia's state-owned publication RT. The US authorities claim the bot farm can all be traced back to one employee. The bot farm consisted of more than 900 social media accounts designed to masquerade as accounts owned by Americans, and their goal was to post a massive amount of information online at once.

It states in the report the accounts posted content about the Ukraine-Russia war, which included videos of Russia's President Vladimir Putin justifying the invasion. The RT employee setup this bot farm by acquiring two domain names from Arizona-based company Namecheap. These domain names were then used to create two email servers that were then used to create nearly 1,000 bot accounts for social media platforms - 968 in total.

The world of AI is being forced upon us whether we like it or not, and as much bragging as these tech giants do, the CEO of Google DeepMind says that the IQ levels of current AI models are not as smart as a domestic cat.

In a recent chat with Tony Blair, the ex-Prime Minister of Britain, Google DeepMind CEO Demis Hassabis compared artificial intelligence versus cat IQ during the Future of Britain Conference 2024, organized by the Institute for Global Change.

Hassabis talked about his work not being focused on AI but rather on AGI (AI = artificial intelligence, AGI = artificial general intelligence), and that is how he is looking at the computer versus cat comparison. Current AI models can write, paint, create music, and more in a human-like fashion, but a domestic house cat has more intelligence. Hassabis said: "At the moment, we're far from human-level intelligence across the board. But in certain areas like games playing (AI is) better than the best people in the world".

Wells Fargo is projecting that AI power demand will surge by an incredible 8050% by 2030, from 8TWh in 2024 up to an incredible 652TWh in 2030.

The world of AI is exploding in more ways than one, with the power consumption of today's high-end AI GPUs like the NVIDIA H100 AI GPU where the H100 in SXM form factor using 700W of power alone. Moving over to NVIDIA's next-generation B200 AI GPU, which will use up to 1200W of power per AI GPU, we can see how these AI power demands are going to get out of control.

AMD's previous-gen Instinct MI250 AI accelerators draw a peak of 560W of power, while the new MI300X AI accelerator from the company consumes 750W at peak, a 50% increase gen-over-gen. Intel doesn't get out of this either, with its new Gaudi 2 AI accelerator using 600W, its new Gaudi 3 AI accelerator uses 900W of power at peak (another 50% increase).

SK hynix is looking for semiconductor engineers to help fill 48 positions related to HBM chips, including manufacturing process engineers to help increase yields and improve testing, FinFET transistor experts, and former Samsung staffers.

In a new report from The Korea Economic Daily, dominance in the AI and HBM space is heating up between South Korean rivals Samsung SK hynix, particularly in the HBM4 memory chip business. SK hynix designs and produces most of its semiconductors in-house, including HBM memory, but for HBM4, the company is outsourcing manufacturing to a foundry or contract chipmaker, with all signs pointing to TSMC.

On top of that, SK hynix is "aggressively" looking for the top talent in the industry -- mostly in South Korea, where Samsung and SK hynix are battling it out -- to advance its HBM technology and oversee tech outsourcing. SK hynix is aiming for the throat, looking at hiring as many Samsung engineers as it can.

South Korean giants Samsung Electronics and SK Group are "speeding up efforts" to secure future dominance in glass substrates, which would be a "game-changer" for the semiconductor and AI industries.

In a new report from the Korea Herald, Samsung and SK Group are reportedly speeding up the development of glass substrate semiconductor research, which could "drastically boost the data capacity and speed of semiconductors in the era of artificial intelligence."

Samsung and SK Group leaders have visited their respective business sites that are producing glass substrates, renewing their commitment to the game-changing semiconductor technology. Glass substrates will overcome the limits of conventional plastic substrates, heavily boosting the performance and power efficiency of future-gen semiconductors.