SecurityBrief US - Technology news for CISOs & cybersecurity decision-makers
United States

Graphics Processing Unit (GPU) Stories

American Stories

Graphics Processing Unit (GPU) stories

Backbase banking AI beats GPT-4.1 in live deployment
Fintech

Backbase banking AI beats GPT-4.1 in live deployment

In a seven-month test, the specialist model answered more banking queries than GPT-4.1 while refusing uncertain prompts less often than expected.

By Karen Joy Bacudo • 4 min read • Last month

Global Stories

Graphics Processing Unit (GPU) stories

Apple unveils Mac Studio with M5 chips for AI workloads
Semiconductors

Apple unveils Mac Studio with M5 chips for AI workloads

New AI tools and faster media workflows are the focus as the updated desktop adds M5 Max and M5 Ultra chips plus Wi-Fi 7.

By Sean Mitchell • 5 min read • Yesterday
Apple unveils M6 & M5 Ultra chips for Mac desktops
Semiconductors

Apple unveils M6 & M5 Ultra chips for Mac desktops

Mac users tackling AI and graphics-heavy work will get more local muscle, as Apple's latest desktop silicon pushes memory and compute higher.

By Sean Mitchell • 4 min read • Yesterday
Apple's Unified Memory powers Kiraa AI's rapid automation
Software-as-a-Service

Apple's Unified Memory powers Kiraa AI's rapid automation

Australian firms can speed month-end processing from days to an hour by running Kiraa AI on Macs instead of costly cloud servers.

By Anthony Caruana • 4 min read • Yesterday
NVIDIA launches NVLink Fusion for custom AI factories
DataCentre infrastructure

NVIDIA launches NVLink Fusion for custom AI factories

It lets cloud operators pair custom XPUs with NVIDIA's rack-scale systems, aiming to cut delays and integration costs in AI build-outs.

By Sean Mitchell • 5 min read • Yesterday
NVIDIA says Vera Rubin boosts AI throughput per megawatt
Data centers

NVIDIA says Vera Rubin boosts AI throughput per megawatt

Energy-constrained AI operators could run far more agentic tasks after NVIDIA said Vera Rubin systems deliver up to 30 times higher throughput per megawatt.

By Sean Mitchell • 4 min read • Yesterday
BrainChip links Akida processors with IBM Symphony
Private Cloud

BrainChip links Akida processors with IBM Symphony

Enterprises could cut GPU use and power costs as the open-source bundle lets Akida neuromorphic chips share IBM's scheduler for inference tasks.

By Joseph Gabriel Lagonsin • 4 min read • Last week
Cloudera adds GPU acceleration to Apache Spark 4.1 pipelines
Private Cloud

Cloudera adds GPU acceleration to Apache Spark 4.1 pipelines

The new feature could cut Spark job runtimes by up to 4x, easing cloud bills for firms running data-heavy AI and analytics workloads.

By Sean Mitchell • 4 min read • Last week
NVIDIA launches Linux GeForce NOW app with cloud tweaks
Virtualisation

NVIDIA launches Linux GeForce NOW app with cloud tweaks

Linux gamers will get smoother access to streamed titles, as Nvidia moves GeForce NOW out of beta and trims cloud gaming lag.

By Mark Tarre • 3 min read • This month
Centuria & ResetData expand AI pipeline beyond 250MW
Semiconductors

Centuria & ResetData expand AI pipeline beyond 250MW

Customer demand is outstripping near-term AI compute capacity as the pair line up power, data halls and GPUs to speed deployments.

By Mark Tarre • 4 min read • This month
Smart stuff in Dell's simplified approach to the venerable PC
Supply Chain

Smart stuff in Dell's simplified approach to the venerable PC

Businesses need not necessarily rush to replace working PCs for AI, as Dell says those with accelerators can still handle workloads.

By Donovan Jackson • 5 min read • This month
Google Cloud launches device platform for app testing
Software-as-a-Service

Google Cloud launches device platform for app testing

The pay-per-minute service could cut testing costs for teams that need to check apps across many handset types and screen sizes.

By Sean Mitchell • 4 min read • This month
Mirendil taps Google Cloud AI hypercomputer for research
Platform-as-a-Service

Mirendil taps Google Cloud AI hypercomputer for research

The startup gains faster access to scarce AI compute as it prepares to scale model training across both TPU and Nvidia systems.

By Mark Tarre • 3 min read • This month
Runware launches modular AI inference pods in weeks
Gaming

Runware launches modular AI inference pods in weeks

The modular units aim to ease AI compute bottlenecks by bringing 1 megawatt of capacity online in weeks, not years.

By Joseph Gabriel Lagonsin • 4 min read • This month
Sharon AI signs USD $373 million cloud deal in Australia
Data centers

Sharon AI signs USD $373 million cloud deal in Australia

Most of Sharon AI's Australian AI Factory capacity is now spoken for, with revenue from the USD $373 million deal due from early 2027.

By Sean Mitchell • 3 min read • This month
Researchers find 36,872 exposed server management interfaces
Ransomware

Researchers find 36,872 exposed server management interfaces

Exposed BMCs can let attackers take persistent control of servers, with nearly 25,000 systems leaking password data before login.

By Sean Mitchell • 4 min read • Last month
Tensormesh & AMD team up to ease AI memory pressure
Data centers

Tensormesh & AMD team up to ease AI memory pressure

The deal could let enterprises serve more AI models on fewer GPUs by shifting cache data into DRAM, SSDs and remote storage.

By Sean Mitchell • 4 min read • Last month
Vultr adds AMD MI455X GPU & Helios rackscale support
Public Cloud

Vultr adds AMD MI455X GPU & Helios rackscale support

Customers facing soaring AI compute demands will get access to one of the first cloud offerings of AMD's latest large-memory GPU and rackscale system.

By Sean Mitchell • 4 min read • Last month