Lifeboat Foundation

Twelve hacktivist group targets Russian entities with destructive cyber attacks, using public tools for maximum damage without financial gain.

A massive infostealer malware operation encompassing thirty campaigns targeting a broad spectrum of demographics and system platforms has been uncovered, attributed to a cybercriminal group named “Marko Polo.”

The threat actors use a variety of distribution channels, including malvertising, spearphishing, and brand impersonation in online gaming, cryptocurrency, and software, to spread 50 malware payloads, including AMOS, Stealc, and Rhadamanthys.

According to Recorded Future’s Insikt Group, which has been tracking the Marko Polo operation, the malware campaign has impacted thousands, with potential financial losses in the millions.

Our final estimate of the achievable inter data center bandwidth by 2030 is 4 to 20 Pbps, which would allow for training runs of 3e29 to 2e31 FLOP. In light of this, bandwidth is unlikely to be a major constraint for a distributed training run compared to achieving the necessary power supply in the first place.

Expanding bandwidth capacity for distributed training networks presents a relatively straightforward engineering challenge, achievable through the deployment of additional fiber pairs between data centers. In the context of AI training runs potentially costing hundreds of billions of dollars, the financial investment required for such bandwidth expansion appears comparatively modest.⁴⁴

We conclude that training runs in 2030 supported by a local power supply could likely involve 1 to 5 GW and reach 1e28 to 3e29 FLOP by 2030. Meanwhile, geographically distributed training runs could amass a supply of 2 to 45 GW and achieve 4 to 20 Pbps connections between data center pairs, allowing for training runs of 2e28 to 2e30 FLOP.⁴⁵ All in all, it seems likely that training runs between 2e28 to 2e30 FLOP will be possible by 2030.⁴⁶ The assumptions behind these estimates can be found in Figure 3 below.

In recent years, these technological limitations have become far more pressing. Deep neural networks have radically expanded the limits of artificial intelligence—but they have also created a monstrous demand for computational resources, and these resources present an enormous financial and environmental burden. Training GPT-3, a text predictor so accurate that it easily tricks people into thinking its words were written by a human, costs $4.6 million and emits a sobering volume of carbon dioxide—as much as 1,300 cars, according to Boahen.

With the free time afforded by the pandemic, Boahen, who is faculty affiliate at the Wu Tsai Neurosciences Institute at Stanford and the Stanford Institute for Human-Centered AI (HAI), applied himself single mindedly to this problem. “Every 10 years, I realize some blind spot that I have or some dogma that I’ve accepted,” he says. “I call it ‘raising my consciousness.’”

This time around, raising his consciousness meant looking toward dendrites, the spindly protrusions that neurons use to detect signals, for a completely novel way of thinking about computer chips. And, as he writes in Nature, he thinks he’s figured out how to make chips so efficient that the enormous GPT-3 language prediction neural network could one day be run on a cell phone. Just as Feynman posited the “quantum supremacy” of quantum computers over traditional computers, Boahen wants to work toward a “neural supremacy.”

Mathematician Bernhard Riemann was born #OTD in 1826.

Bernhard Riemann was another mathematical giant hailing from northern Germany. Poor, shy, sickly and devoutly religious, the young Riemann constantly amazed his teachers and exhibited exceptional mathematical skills (such as fantastic mental calculation abilities) from an early age, but suffered from timidity and a fear of speaking in public. He was, however, given free rein of the school library by an astute teacher, where he devoured mathematical texts by Legendre and others, and gradually groomed himself into an excellent mathematician. He also continued to study the Bible intensively, and at one point even tried to prove mathematically the correctness of the Book of Genesis.

Although he started studying philology and theology in order to become a priest and help with his family’s finances, Riemann’s father eventually managed to gather enough money to send him to study mathematics at the renowned University of Göttingen in 1846, where he first met, and attended the lectures of, Carl Friedrich Gauss. Indeed, he was one of the very few who benefited from the support and patronage of Gauss, and he gradually worked his way up the University’s hierarchy to become a professor and, eventually, head of the mathematics department at Göttingen.

New Android malware ‘Ajina. Banker’ targets bank customers in Central Asia, stealing financial data and intercepting 2FA messages via Telegram channels.

SparkLabs — an early-stage venture capital firm that has made a name for itself for backing OpenAI as well as a host of other AI startups such as Vectara, Allganize, Kneron, Anthropic, xAI, Glade (YC S23) and Lucidya AI — is gearing up to double down on more startups in the space. The VC firm announced Tuesday that it has closed a new $50 million fund, AIM AI Fund, which will back AI startups out of its own AIM-X accelerator in Saudi Arabia as well as other AI startups across the globe.

SparkLabs’ new fund and its wider investment aims underscore the bigger trends that have swirled around artificial intelligence for the last few years. The explosion of interest in generative AI in particular has led to a surge of startups in the space, as well as a rush of investors looking for the next Open AI — or at the very least, a startup that a bigger company might snap up as it looks to sharpen its own AI edge.

It also points to how the AI opportunity continues to widen beyond Silicon Valley. AIM-X is an AI-focused startup accelerator that SparkLabs launched earlier this year in the kingdom as part of its AI Mission, a national initiative to bolster AI technology over the next five years.

Finally, the goal of any healthcare organization is to provide the best possible care to patients. Predictive AI can contribute significantly to this goal by enabling more accurate diagnoses, tailored treatment plans and earlier interventions.

From the patient’s perspective, this translates to better health outcomes, reduced hospital stays and increased satisfaction with their care. For healthcare organizations, improved patient experiences lead to higher patient retention rates, positive word-of-mouth referrals and better performance on patient satisfaction metrics, which are increasingly tied to reimbursement rates in many healthcare systems.

As we’ve explored, the benefits of predictive AI extend far beyond improved diagnostics and treatment plans. It’s a catalyst for operational excellence, financial optimization, availability of investments and long-term growth. From resource management to building an authoritative brand, predictive AI touches every aspect of the healthcare business environment.

“For equities, all attention is now on Nvidia’s earnings release tonight, which has helped to drive significant moves recent quarters,” Deutsche Bank strategists said in a note Wednesday morning. “Bear in mind that Nvidia’s share price is already up +159% on a YTD basis, making it the top performer in the entire S&P 500, and it has risen by more than +1000% since its low in October 2022.”

Traders are also waiting on comments from Atlanta Fed President Raphael Bostic, who’s scheduled to speak after the closing bell. His remarks could provide more guidance on the path of Fed rate cuts this year, with investors pricing in as many as 150 basis points worth of cuts by year-end, according to the CME FedWatch tool.