Lifeboat Foundation

The new AI strategy, which includes the construction of a supercomputer, will cost the UK £900 million ($1.2 billion).

The United Kingdom (U.K.) has announced plans to develop its own ChatGPT version, “BritGPT” as part of a new artificial intelligence (AI) strategy.

“These investments will provide scientists with access to cutting-edge computing power and bring a significant uplift in computing capacity to the AI community,” reads the Spring Budget 2023 plan.

Continue reading “‘BritGPT’: UK plans ChatGPT-like superpower to counter China’s AI influence” »

No, it’s not hypermodern art. This image, generated by NASA

Established in 1958, the National Aeronautics and Space Administration (NASA) is an independent agency of the United States Federal Government that succeeded the National Advisory Committee for Aeronautics (NACA). It is responsible for the civilian space program, as well as aeronautics and aerospace research. Its vision is “To discover and expand knowledge for the benefit of humanity.” Its core values are “safety, integrity, teamwork, excellence, and inclusion.” NASA conducts research, develops technology and launches missions to explore and study Earth, the solar system, and the universe beyond. It also works to advance the state of knowledge in a wide range of scientific fields, including Earth and space science, planetary science, astrophysics, and heliophysics, and it collaborates with private companies and international partners to achieve its goals.

Its Up!

We’ve created GPT-4, the latest milestone in OpenAI’s effort in scaling up deep learning. GPT-4 is a large multimodal model (accepting image and text inputs, emitting text outputs) that, while less capable than humans in many real-world scenarios, exhibits human-level performance on various professional and academic benchmarks. For example, it passes a simulated bar exam with a score around the top 10% of test takers; in contrast, GPT-3.5’s score was around the bottom 10%. We’ve spent 6 months iteratively aligning GPT-4 using lessons from our adversarial testing program as well as ChatGPT, resulting in our best-ever results (though far from perfect) on factuality, steerability, and refusing to go outside of guardrails.

Over the past two years, we rebuilt our entire deep learning stack and, together with Azure, co-designed a supercomputer from the ground up for our workload. A year ago, we trained GPT-3.5 as a first “test run” of the system. We found and fixed some bugs and improved our theoretical foundations. As a result, our GPT-4 training run was (for us at least!) unprecedentedly stable, becoming our first large model whose training performance we were able to accurately predict ahead of time. As we continue to focus on reliable scaling, we aim to hone our methodology to help us predict and prepare for future capabilities increasingly far in advance—something we view as critical for safety.

Continue reading “GPT-4” »

Now its building one that even bigger and even more sophisticated.

Nearly five years ago, a little-known company approached Microsoft with a special request to put together computing horsepower to the scale it had never done before. Microsoft then spent millions of dollars in putting together tens of thousands of powerful chips to build a supercomputer. OpenAI used this to train its large language model, GPT, and the rest, as they say, is history.

Microsoft is no stranger to building artificial intelligence (AI) models that help users work more efficiently. The automatic spell checker that has helped millions of users is an example of an AI model trained on language.

Continue reading “Microsoft spent millions to put together a supercomputer for OpenAI” »

Physicists have created a novel type of analog quantum computer capable of addressing challenging physics problems that the most powerful digital supercomputers cannot solve.

A groundbreaking study published in Nature Physics.

As the name implies, Nature Physics is a peer-reviewed, scientific journal covering physics and is published by Nature Research. It was first published in October 2005 and its monthly coverage includes articles, letters, reviews, research highlights, news and views, commentaries, book reviews, and correspondence.

The new supercomputer system can predict the occurrence of linear rainbands, which are clouds that trigger heavy rain, leading to natural disasters.

Trust Japan to get a supercomputer to predict heavy rain and other natural disasters like landslides and flooding.

Japan has always had to deal with natural disasters as the island is located along an area where several tectonic plates meet. The country is highly vulnerable to natural disasters such as earthquakes, tsunamis, and natural disasters. These have only been exacerbated due to climate change.

“Computers that run on this ‘biological hardware’ could in the next decade begin to alleviate energy-consumption demands of supercomputing.”

Johns Hopkins University researchers have outlined plans for a “bio-computer” that is highly feasible in our lifetime.

“Computing and artificial intelligence have been driving the technology revolution, but they are reaching a ceiling,” Thomas Hartung, a professor of environmental health sciences at the Johns Hopkins Bloomberg School of Public Health and Whiting School of Engineering, who is spearheading the work, said in a statement.

But the computing power necessary for a company to adopt in-house AI capabilities is enormous, and that’s where Nvidia’s new service offering comes in. Dubbed “DGX Cloud,” Nvidia is offering an AI supercomputer accessible to its customers via a web browser. The company partnered with various cloud providers, including Microsoft, Google, and Oracle, to launch the service.

“Nvidia AI as a service offers enterprises easy access to the world’s most advanced AI platform, while remaining close to the storage, networking, security and cloud services offered by the world’s most advanced clouds,” Huang explained.

“Nvidia AI is essentially the operating system of AI systems today,” Huang also said.

Quantum computers are highly energy-efficient and extremely powerful supercomputers. But for these machines to realize their full potential in new applications like artificial intelligence or machine learning, researchers are hard at work at perfecting the underlying electronics to process their calculations. A team at Fraunhofer IZM are working on superconducting connections that measure a mere ten micrometers in thickness, moving the industry a substantial step closer to a future of commercially viable quantum computers.

With the extreme computing power they promise, quantum computers have the potential to become the driving force for technological innovations in all areas of modern industry. By contrast with the run-of-the-mill computers of today, they do not work with bits, but with qubits: No longer are these units of information restricted to the binary states of 1 or 0.

With quantum superposition or entanglement added, qubits mean a great leap forward in terms of sheer speed and power and the complexity of the calculations they can handle. One simple rule still holds, though: More qubits mean more speed and more computing power.