Amazon's Zoox robotaxi service offers free driverless rides to Las Vegas Strip locations including Resorts World, AREA15 and Topgolf as part of its public launch.

With its focus on innovative products and cutting-edge technology, the annual Consumer Electronics Show (CES) has not historically paid much attention to energy companies. But there were signs of a shift at this year's Las Vegas event, as the tech sector begins to confront its substantial energy needs, which are certain to grow as cloud computing and artificial intelligence advance. "If you'd asked me to do CES five years ago, I wouldn't necessarily have seen the point," said Sebastien Fiedorow, chief executive of the French start-up Aerleum, which manufactures synthetic fuel from carbon dioxide.

The holiday season has barely begun, but some of us are already getting ready for CES 2025. Shortly after New Year's Day, many from the Engadget team will be packing our bags to fly to Las Vegas, where we'll be covering tech's biggest annual conference. As usual, our inboxes are already flooded with pitches from companies that are planning to be there, and our calendars are filling up with appointments for briefings and demos. Based on our experience, as well as observation of recent industry trends, it's fairly easy to make educated predictions about what we might see in January. Over the years, the focus of the conference has spanned areas like TVs, cars, smart home products and personal health, with a smattering of laptops and accessories thrown in.

The frenzied and intoxicating showcase for consumer technology known as CES took place this week in Las Vegas. Every January, the industry's big shindig descends on this city in the Nevada desert, drawing tech manufacturers, retailers, distributors, members of the press, gadget fans, and regular old lookie-loos into the fray. The Las Vegas Convention Center, hotel expos halls, nightclubs, restaurants, and event centers are stuffed with talking screens, self-driving cars, flying cars, self-adjusting audio speakers, and ChatGPT-enabled appliances for the smart home. Indeed, this is the year that AI ate everything in sight; old products were freshened up by an injection of machine intelligence, and new products were launched to help people interface with these new generative tools. Our photographer Alex Welsh captured some of this consumer tech revolution in full swing as he roamed the halls of CES 2024.

New virtual machines for Microsoft Azure allow developers to create generative AI apps that can be scaled to work with thousands of Nvidia H100 GPUs. The ND H100 v5 VM series on Azure, which works in tandem with Quantum-2 InfiniBand networking, boosts the performance of large-scale deployments by companies such as OpenAI, creators of the much talked about ChatGPT, and Nvidia's chips. The new supercomputing system in the cloud provides the type of infrastructure required to handle the latest large-scale AI training models, according to Matt Vegas, principal product manager for Azure HPC and AI at Microsoft. "Generative AI applications are rapidly evolving and adding unique value across nearly every industry," Vegas wrote in a blog post this week. "From the newly released AI-powered Bing and Edge to AI-powered assistance in Microsoft Dynamics 365, AI is becoming a pervasive component of software and how we interact with it. We want to ensure that our AI infrastructure will be there to pave the way."

Simulating physical network paths (e.g., Internet) is a cornerstone research problem in the emerging sub-field of AI-for-networking. We seek a model that generates end-to-end packet delay values in response to the time-varying load offered by a sender, which is typically a function of the previously output delays. The problem setting is unique, and renders the state-of-the-art text and time-series generative models inapplicable or ineffective. We formulate an ML problem at the intersection of dynamical systems, sequential decision making, and time-series modeling. We propose a novel grey-box approach to network simulation that embeds the semantics of physical network path in a new RNN-style model called RBU, providing the interpretability of standard network simulator tools, the power of neural models, the efficiency of SGD-based techniques for learning, and yielding promising results on synthetic and real-world network traces.

Estimation of riverbed profiles, also known as bathymetry, plays a vital role in many applications, such as safe and efficient inland navigation, prediction of bank erosion, land subsidence, and flood risk management. The high cost and complex logistics of direct bathymetry surveys, i.e., depth imaging, have encouraged the use of indirect measurements such as surface flow velocities. However, estimating high-resolution bathymetry from indirect measurements is an inverse problem that can be computationally challenging. Here, we propose a reduced-order model (ROM) based approach that utilizes a variational autoencoder (VAE), a type of deep neural network with a narrow layer in the middle, to compress bathymetry and flow velocity information and accelerate bathymetry inverse problems from flow velocity measurements. In our application, the shallow-water equations (SWE) with appropriate boundary conditions (BCs), e.g., the discharge and/or the free surface elevation, constitute the forward problem, to predict flow velocity. Then, ROMs of the SWEs are constructed on a nonlinear manifold of low dimensionality through a variational encoder. Estimation with uncertainty quantification (UQ) is performed on the low-dimensional latent space in a Bayesian setting. We have tested our inversion approach on a one-mile reach of the Savannah River, GA, USA. Once the neural network is trained (offline stage), the proposed technique can perform the inversion operation orders of magnitude faster than traditional inversion methods that are commonly based on linear projections, such as principal component analysis (PCA), or the principal component geostatistical approach (PCGA). Furthermore, tests show that the algorithm can estimate the bathymetry with good accuracy even with sparse flow velocity measurements.

Las Vegas's search for an adaptive security solution led it to deploy Darktrace AI across its enterprise, cloud and industrial networks. Background In recent years, Las Vegas has become a prototypical Smart City. As riders glide down the Strip aboard the first completely autonomous shuttle ever deployed on a public roadway, they are unlikely to notice much trash on the sidewalk – the city's surveillance cameras stream to an AI service that directs clean-up crews towards concentrations of litter. And when rush hour approaches, its passengers can rest assured that an array of connected sensors are helping officials anticipate gridlock at busy intersections. But while smart infrastructure enables Las Vegas to achieve new heights of efficiency, conventional security tools are largely ill-equipped to defend the hybrid cloud and industrial networks that power this infrastructure.

The Las Vegas self-driving shuttle is one of many smart cities projects. Welcome to Las Vegas, city of smart lights, self-driving shuttles and startups. Away from the glittering, casino-strewn area known as the Strip is a far more pedestrian-looking area. It's just a 15-minute drive from Las Vegas Boulevard, but it feels like a different world. It's quiet downtown, because while the Strip was thronging with 200,000 extra visitors for CES 2020 last week, the streets here were cold and empty.