Tag: gpus

LeftoverLocals GPU Flaw Exposes AI Data in Devices Equipped with Apple, AMD, and Qualcomm Hardware

A security flaw affecting GPUs from four hardware manufacturers that exposed artificial intelligence (AI) data was unearthed by security researchers. The issue impacts several devices equipped with GPUs from these firms, including some iPhone, iPad, and Mac computers. Hackers can exfiltrate personal information being used in AI operations on the local memory of affected devices — including large language models (LLMs) used by services like Google, Meta, ChatGPT maker OpenAI, and Microsoft using a few lines of code, according to researchers.

Researchers at Trail of Bits uncovered a security flaw affecting GPUs from AMD, Apple, Imagination, and Qualcomm that has been dubbed LeftoverLocals. This vulnerability is related to the affected device’s GPU and allows hackers to access information via local memory created by another process. Arm, Intel, and Nvidia GPUs are reportedly unaffected by the same security flaw.

In a detailed disclosure published earlier this week, the researchers highlight how the security flaw affects LLMs and machine learning (ML) models that are run on impacted devices. They were able to build a proof of concept (PoC) of the attack that allowed them to access information from another user’s LLM session that was being run in a different process.

A demonstration of an attacker listening in on an interactive LLM chat session
Photo Credit: Screenshot/ Trail of Bits

 

By running a few lines of code, a hacker can use the LeftoverLocals security flaw to reconstruct the LLM response in an interactive session “with high precision”, according to the researchers. The flaw was discovered by Tyler Sorensen and is being tracked by CVE-2023-4969.

The researchers state that they reached out to Apple and received a response on January 13, while the company has patched some devices with the A17 Pro — that powers the iPhone 15 Pro and 15 Pro Max — and M3 chip series, but other devices have not been patched, such as the M2-powered MacBook Air.

Meanwhile, AMD has stated is still exploring ways to mitigate the security vulnerability and Qualcomm has issued a patch with its v2.07 firmware that fixes the flaw on some devices, while others could still remain impacted. Affected Imagination GPUs were patched last month as part of the recent DDK 23.3 release, according to the researchers.

Affiliate links may be automatically generated – see our ethics statement for details.

Check out our Latest News and Follow us at Facebook

Original Source

Nvidia GeForce RTX 4060 Ti, GeForce RTX 4060 With Ray Tracing, DLSS 3 Support Launched in India

Nvidia GeForce RTX 4060 and GeForce RTX 4060 Ti were launched by the company on Thursday as the company’s latest graphics cards built on the current-gen Ada Lovelace architecture. The more powerful GeForce RTX 4060 Ti is available in 8GB and 16GB memory configurations, while the RTX 4060 model is available with 8GB of memory. Both models support Nvidia’s deep learning super sampling (DLSS) for over 300 games as well as upcoming titles such as The Lord of the Rings: Gollum and Diablo IV.

The company says that the new GeForce RTX 4060 Ti graphics card is 1.7 times faster than the GeForce RTX 3060 Ti model. It is equipped with 8GB or 16GB of GDDR6 video memory, along with 32MB of L2 cache. Meanwhile, the GeForce RTX 4060 model has 24MB of L2 cache and is available in a single 8GB memory configuration.

The company says these chips are aimed at gamers, video editors, 3D artists, and other content creators such as broadcasters, who will see 40 percent better encoding efficiency, with the eighth NVENC video encoder from Nvidia. Similarly, 3D artists will see 45 percent faster performance over last year’s RTX 3060 GPU series, according to Nvidia.

Pricing for the Nvidia GeForce RTX 4060 Ti starts at Rs. 41,000 for the model with 8GB of video memory, and will go on sale on May 24. The RTX 4060 Ti 16GB variant is priced at Rs. 51,000 and will go on sale in July, along with the RTX 4060 GPU with 8GB of memory, which is priced at Rs. 31,000. For comparison, pricing for the Nvidia GeForce RTX 4070 Ti starts at Rs. 80,000 while the RTX 4070 starts at Rs. 62,000.

Nvidia says that the new GPUs based on the company’s Ada Lovelace architecture are equipped with third-generation ray tracing (RT) cores and fourth-generation Tensor cores — the latter are aimed at speeding up artificial intelligence tasks, including Nvidia’s own framerate-multiplying DLSS 3 technology. The firm says it has added improvements like shader execution reordering, opacity micromap, and displaced micro-mesh engines to improve ray tracing performance over the previous generation.

Earlier this year, Nvidia announced its GeForce RTX 4070 Ti desktop GPU in January at CES 2023, months after the company “unlaunched” its GeForce RTX 4080 (12GB) model following negative press and consumer protests. The new GeForce RTX 4070 Ti GPU, which is also based on the Ada Lovelace architecture like the GeForce RTX 4060 series, had a much lower price at launch compared to the previous product that was “unlaunched” by the company.

Google I/O 2023 saw the search giant repeatedly tell us that it cares about AI, alongside the launch of its first foldable phone and Pixel-branded tablet. This year, the company is going to supercharge its apps, services, and Android operating system with AI technology. We discuss this and more on Orbital, the Gadgets 360 podcast. Orbital is available on Spotify, Gaana, JioSaavn, Google Podcasts, Apple Podcasts, Amazon Music and wherever you get your podcasts.
Affiliate links may be automatically generated – see our ethics statement for details.

Check out our Latest News and Follow us at Facebook

Original Source

Exit mobile version