2024 is witnessing a remarkable shift in the landscape of generative AI. While cloud-based models like GPT-4 continue to evolve, running powerful generative AI directly on local devices is becoming increasingly viable and attractive. This local execution of generative AI can transform how small businesses, developers, and everyday users benefit from AI. Let's explore the critical aspects of this exciting trend.

Traditionally, generative AI has relied on cloud services for its computational power. Although the cloud has driven significant innovation, it faces several challenges in deploying generative AI applications. Increasing data breaches have heightened concerns about keeping sensitive information secure. Processing data locally with on-device AI minimizes exposure to external servers.

Cloud-based AI also needs help with latency issues, leading to slower responses and a less smooth user experience. On-device AI can significantly reduce latency, providing faster responses and a smoother experience, which is particularly crucial for real-time applications like autonomous vehicles and interactive virtual assistants.

Another critical challenge for cloud-based AI is sustainability. Data centers, the backbone of cloud computing, are notorious for high energy consumption and a substantial carbon footprint. As the world grapples with climate change, reducing technology's environmental impact has become paramount. Local generative AI offers a compelling solution, reducing reliance on energy-intensive data centers and minimizing the need for constant data transfers.

Cost is another significant factor. While cloud services are robust, they can be expensive, especially for continuous or large-scale AI operations. By harnessing the power of local hardware, companies can reduce operational costs, which is particularly beneficial for smaller businesses and startups that may find cloud computing costs prohibitive.

Additionally, continuous dependency on an internet connection is a significant drawback of cloud-based AI. On-device AI eliminates this dependency, allowing uninterrupted functionality even in areas with poor or no internet connectivity. This aspect is particularly advantageous for mobile applications and remote or rural areas where internet access may be unreliable.

We witness a remarkable transformation towards local generative AI as these factors converge. This shift promises enhanced performance, improved privacy, and greater democratization of AI technology, making powerful tools available to a broader audience without the need for constant internet connectivity.

Besides the challenges of cloud-powered generative AI, integrating AI capabilities directly into mobile devices is emerging as a pivotal trend in recent years. Mobile phone manufacturers increasingly invest in dedicated AI chips to enhance performance, efficiency, and user experience. Companies like Apple with its A-series chips, Huawei with its Ascend AI processor, Samsung with its Exynos lineup, and Qualcomm with its Hexagon neural processing units are leading this charge.

Neural Processing Units (NPUs) are emerging as specialized AI processors designed to implement generative AI on mobile devices. These brain-inspired processors handle complex AI tasks efficiently, enabling faster and more accurate data processing directly on mobile devices. Integrated with other processors, including CPU and GPU, into their SoCs (System-on-a-Chip), NPUs efficiently cater to the diverse computational needs of generative AI tasks. This integration allows generative AI models to run more smoothly on the device, enhancing the overall user experience.

The rising integration of generative AI into everyday applications, such as Microsoft Office or Excel, has given rise to AI PCs. Significant advancements in AI-optimized GPUs support this emergence. Initially designed for 3D graphics, graphical processing units (GPUs) have proven remarkably effective at running neural networks for generative AI. As consumer GPUs advance for generative AI workloads, they also become increasingly capable of handling advanced neural networks locally. For instance, the Nvidia RTX 4080 laptop GPU, released in 2023, leverages up to 14 teraflops of power for AI inference. As GPUs become more specialized for ML, local generative AI execution will scale significantly in the coming days.

AI-optimized operating systems support this development by dramatically speeding up the processing of generative AI algorithms while seamlessly integrating these processes into the user's everyday computing experience. Software ecosystems have been evolving to leverage generative AI capabilities, with AI-driven features such as predictive text, voice recognition, and automated decision-making becoming core aspects of the user experience.

The implications of this technological leap are profound for both individual consumers and enterprises. For consumers, the appeal of AI PCs is substantial due to their convenience and enhanced functionality. For enterprises, the potential of AI PCs is even more significant. Licensing AI services for employees can be costly, and legitimate concerns about sharing data with cloud AI platforms exist. AI PCs offer a cost-effective and secure solution to these challenges, allowing businesses to integrate AI capabilities directly into their operations without relying on external services. This integration reduces costs and enhances data security, making AI more accessible and practical for workplace applications.

Generative AI is rapidly transforming industries across the globe. Edge computing brings data processing closer to devices, reducing latency and enhancing real-time decision-making. The synergy between generative AI and edge computing allows autonomous vehicles to interpret complex scenarios instantly and intelligent factories to optimize production lines in real-time. This technology empowers next-generation applications, such as smart mirrors providing personalized fashion advice and drones analyzing crop health in real-time.

According to a report, over 10,000 companies building on the NVIDIA Jetson platform can now leverage generative AI to accelerate industrial digitalization. The applications include defect detection, real-time asset tracking, autonomous planning, human-robot interactions, and more. ABI Research predicts that generative AI will add $10.5 billion in revenue for manufacturing operations worldwide by 2033. These reports underscore the crucial role that local generative AI will increasingly play in driving economic growth and fostering innovation across various sectors shortly.

The convergence of local generative AI, mobile AI, AI PCs, and edge computing marks a pivotal shift in harnessing AI's potential. By moving away from cloud dependency, these advancements promise enhanced performance, improved privacy, and reduced costs for businesses and consumers alike. With applications spanning from mobile devices to AI-driven PCs and edge-enabled industries, this transformation democratizes AI and accelerates innovation across diverse sectors. As these technologies evolve, they will redefine user experiences, streamline operations, and drive significant economic growth globally.

Read the rest here:
Local Generative AI: Shaping the Future of Intelligent Deployment - Unite.AI

In the modern day, cloud computing is no longer considered simply as "the Cloud" due to advancements in public and private clouds and, more recently, the creation of the hybrid cloud. In simple terms, the hybrid cloud combines the private cloud and the public cloud in a single IT infrastructure. If executed successfully, businesses are provided with short-term, cost-effective solutions and increased flexibility based on their specific needs.

Companies will adopt this method of cloud deployment for multiple reasons. However, a recentVanguard reportrevealed that the main drivers behind hybrid cloud migration are improving operation scalability (50%), business innovation (46%), and migrating data across IT workloads (42%). Therefore, along with the rapid technological advancements of data and Artificial Intelligence (AI), the hybrid cloud is poised to play a key role in the future of IT infrastructure.

The cloud has come a long way since its inception, moving away from a simple data storage solution. Its creation began as a result of traditional computing consuming significant amounts of power and physical space. Leading public cloud providers offered a cost-effective solution, allowing users to easily scale up or down based on demand. Meanwhile, due to security concerns and the introduction of data privacy laws, the private cloud emerged as a way for businesses to maintain strict control over their data.

Read the Full Article on Network Computing

Visit link:
Is the Hybrid Cloud the New Normal in 2024? - InformationWeek

Source: Blackboard / Shutterstock

Technology is constantly evolving, and each trend has its cycles. Cloud computing gained prominence and popularity at the beginning of the last decade, and this trend reemerged during the Covid-19 pandemic. The moment is approaching when cloud infrastructure stocks will show their multiple growth again.

Reviews by leading investors may mention that further development of artificial intelligence (AI) requires the latest chips. However, these recommendations often overlook the need for adapted high-powered servers that can be provided by cloud market leaders. Further, the infrastructure and relevant data should be kept in reliable and convenient storage. So, cloud solutions come into play, giving individuals, corporations, and government agencies the ability to thrive in the web environment without the hassle of running their servers. Each of the presented stocks has gained 50% or more over the past year and are not ready to rest on its laurels.

In investors minds, Alphabet (NASDAQ:GOOG, NASDAQ:GOOGL) is associated primarily with profits from the advertising business.

Google Search ranks first among the companys successes and accounts for 90% of the market. Search, display, network advertising and YouTube form the basis of the business and provide the company with super-profits. In the first quarter of 2024, Google Search revenue increased by 14.3% year-over-year (YOY), but Alphabet intends to diversify its interests. Google Cloud strengthens the results of the mega-company, as it has added another thousand products and features to the total number. Alphabet is in a state of constant competition with other cloud infrastructure stocks and is improving to withstand the influence of other tech giants.

Alphabet, the third-largest cloud infrastructure platform, is showing positive profitability, encouraging analysts to make optimistic forecasts. The current P/E ratio is close to 28, indicating a low price compared to other tech players. The companys revenue is projected to grow by 11% CAGR through 2026. At the same time, earnings per share (EPS) is expected to grow by 20% on an average annualized basis.

Source: Tada Images / Shutterstock.com

While the general public was looking away at the hype of Nvidia (NASDAQ:NVDA), Amazon (NASDAQ:AMZN) has been climbing step by step to the top of the tech world. The $2 trillion market capitalization is one of its latest achievements, cementing AMZNs position among the top cloud infrastructure stocks.

After a sharp loss of height in 2022, Amazon has shown that it can draw conclusions and catch up with competitors. It made timely staff changes and cut costs in several areas. In October 2023, the price was close to $120, while in the summer of 2024, it reached $197. Such results seem incredible compared to the starting value of 7 cents in 1997, but the stock is ready to surprise investors in the future.

Unlike new names on the market, Amazon has repeatedly proven its adaptability and capacity to thrive in several areas. It is difficult to imagine the modern market of technological solutions without Amazon Web Services (AWS), which steadily holds its position against the competition from Microsoft Azure and Google Cloud. With annual revenues of more than $100 billion, Amazon can conduct researches and be among the first to implement global cloud solutions.

Source: shutterstock.com/LCV

Late 2023 and early 2024 allowed the investment community to see a balanced and innovative company growing. IBM (NYSE:IBM) made a jump in the price per share from $138 to $197, proving the feasibility of diversifying its interests.

The company is generally pursuing a well-thought-out strategy, but several key initiatives have brought it to the top of the tech worlds Olympus. The acquisition of Red Hat for $34 billion in 2019 triggered the growth of the companys hybrid cloud computing capabilities.

IBM recognized the potential of artificial intelligence in time and launched the WatsonX platform. It has become the home of new AI-based solutions and has satisfied a large share of the demand for AI services. With a strong cloud portfolio and backing from tech-focused investors, IBM has set its sights on the field of quantum computing. In collaboration with Fundacin Ikerbasque, the company has entered the forefront of computer science, preparing the ground for unique offerings. Analysts estimates elevate IBM among cloud infrastructure stocks. Despite the current price of about $175.10, Goldman Sachs initiated coverage with a buy rating and a $200 price target.

On the date of publication, Julia Magas did not hold (either directly or indirectly) any positions in the securities mentioned in this article. The opinions expressed in this article are those of the writer, subject to the InvestorPlace.comPublishing Guidelines.

On the date of publication, the responsible editor did not have (either directly or indirectly) any positions in the securities mentioned in this article.

Julia Magas is a writer who covers the latest trends in finance and technology. Her work is published in a number of financial media outlets such as Nasdaq, Cointelegraph, Investing, SeekingAlpha, FXEmpire, and Beincrypto. She primarily covers cryptocurrency and blockchain technology with a focus on market performance, innovations and trends.

Visit link:
Cloud Infrastructure Stocks For Those Investing in The Future - InvestorPlace

Cognizant announced the launch of Cognizant Neuro Edge, a new platform in the Cognizant Neuro suite, designed to empower businesses across industries to leverage artificial intelligence and generative AI at the edge.

Edge computing enables enterprises to access computing power via sensors and devices on their networks, reducing dependency on centralized servers and the cloud. Neuro Edge is designed to power the entire value chain of edge AI, from chips and devices to applications and business solution deployments, shortening the path to business value.

The new platform expands Cognizant's Neuro AI capabilities, offering a hybrid Cloud + Edge AI solution that supports instant decision-making and complements Cloud AI with deeper insights from longitudinal data. The industry agnostic platform offers advantages for industries where data privacy, security, and real-time decision-making are critical.

Neuro Edge facilitates real-time interactions with devices, helping businesses to accelerate decision-making, reduce data costs and privacy risks and maintain operational stability even in low bandwidth scenarios. The platform is cloud agnostic, making it ideal for hybrid-and-multi-cloud environments, with computing power placed directly at the device. Some potential key applications across industries include:

- Healthcare: Assisting doctors in their real-time decision-making by drawing on diagnostic sensors;

- MedTech: Enabling real-time, on device adjustment options and recommendations based on patient data;

- Energy: Optimizing operations in power generation plants; optimizing response to weather events;

- Logistics: Streamlining fleet performance and reducing downtime through in-vehicle data processing;

- Telecommunications: Enhancing network security, resiliency and automation, leading to lower operating costs;

- Manufacturing: Predicting equipment failures to help optimize uptime and operating costs;

- Retail: Enabling intelligent video analysis for real real-time theft prevention and in-store traffic pattern monitoring to enhance customer service;

Transforming driver and passenger experience by enabling real-time, context-aware and private recommendations via cloud connection.

As an example of an application in the automotive industry, Cognizant has been working with Qualcomm Technologies to deploy generative AI at the automotive edge and transform the driving experience.

Neuro Edge integrates APIs with the edge ecosystem, encompassing sensors, silicon vendors, edge devices, and enterprise applications. It offers a foundational architecture and illustrative applications tailored for diverse industrial scenarios. Additionally, it is equipped with monitoring agents that track performance and facilitate ongoing enhancements through feedback loops.

By using Neuro Edge, enterprises will be equipped to create and manage Edge AI applications faster and more easily. The platform helps demystify the intricacies of generative AI, enabling businesses to focus on business value.

Vibha Rustagi, Global Head of IoT and Engineering, Cognizant

Enterprises are increasingly embracing edge computing to enhance the responsiveness of their distributed devices and extract meaningful insights from the data they generate. Cognizant Neuro Edge is a powerful example of Cognizant's leadership in developing a new approach to layering of on-board computing and processing with cloud services, paving the way for businesses to unlock a range of generative AI-driven benefits around operational efficiency, cost and risk reduction.

Nakul Duggal, Group GM, Automotive, Industrial & Embedded IoT and Cloud Computing, Qualcomm Technologies

The automotive industry is going through unprecedented change affecting the entire ecosystem. Our work with Cognizant to extend the Snapdragon Digital Chassis' generative AI capabilities and build a connected services platform creates new opportunities for automakers to develop highly personalized and contextually relevant experiences for both drivers and passengers. We look forward to building new intelligent solutions together that will redefine the in-vehicle experience and drive the future of the automotive industry.

Joel Martin, executive research leader for HFS's technology, media, and telecommunications research at HFS, a leading global research and analysis firm

Implementing AI with edge technologies will be important for firms looking to maximize the impact of their investment in both AI and IoT. I believe Cognizant's Neuro Edge will be an important solution improving functionality, reducing latency, and securing AI at the edge with enterprise-grade capabilities.

Read the rest here:
Cognizant Unveils New Platform to Leverage AI & Generative AI at the Edge - The Fast Mode

Who is responsible for security in the public cloud? This is a question businesses need to consider as they increasingly deploy more workloads and use cloud-based IT infrastructure, platform services and applications.

In Gartners How to make integrated IaaS and PaaS more secure than your own data center report, analysts discuss the benefits of adopting a cloud-native approach to IT security.

Gartner defines a cloud-native mindset as a way to consider IT infrastructure and applications in the cloud as modular and microservices-based. The report authors state that such an architecture is typically container-based, orchestrated and incorporates heavy use of application programming interfaces (APIs). In addition, Gartner says such IT infrastructure is updated using an immutable infrastructure approach.

However, the analysts warn that such an approach does not work well for on-premise IT. The on-premise architectural patterns and their associated tools are poorly suited for the public cloud and will likely frustrate the needs of developers and business units adopting public cloud for its dynamic and ephemeral nature, they note in the report.

Gartner urges IT security leaders responsible for cloud security to be open to embracing new approaches, patterns, products and best practices, and consider alternative IT security technology providers when adopting public cloud.

While largely a benefit, the public cloud also leaves organisations open to public cloud security risks, particularly when they allow users to access on-demand services from various locations using different devices. Beji Jacob, who is on the ISACA emerging trends working group, describes cloud security as technology and techniques engineered to prevent and mitigate threats to an organisations cyber security.

Companies must implement cloud computing security to support both digital transformations and the use of cloud-based tools to protect assets, he says, adding that cloud security works by combining several technologies, all designed to tighten cyber defences for off-premise data and applications.

Rob Dartnall, CEO of SecAlliance, regularly conducts threat-led penetration tests (TLPT) that are part of regulatory frameworks, such as the Bank of Englands CBEST targeted assessment and the UK governments intelligence-led simulated attack framework, GBEST, in the UK.

A key component of the threat intelligence element of these tests is called targeting intelligence, he says. Essentially, it is hostile reconnaissance of an entity that includes many things, but importantly, the reconnaissance of the perimeter and cloud services of an entity to look for weaknesses that could be used to gain a foothold.

In Dartnalls experience, although technical exploitation of a perimeter service by the red teamer is rare against mature entities such as banks, the discovery of shadow services, intellectual property ranges and domains that the entity was not aware of is certainly not rare.

He says there is a direct correlation between those entities that suffer a breach and those that have deployed external attack surface management (EASM). This is an approach to perimeter security where an internal team or external security service provider continuously looks at the perimeter and beyond, not only looking at what is running, versions, services and ports, security controls and misconfigurations, but also at new shadow services, usually accidentally set up by rogue developers, engineers or architects. These shadow IT services, he says, consistently lead to security incidents and data breaches.

There is a role for artificial intelligence (AI) and machine learning (ML), which can operate at a large scale, utilising learning, and can adapt to an organisations data protection needs. By increasing automation, decision-making can be sped up, and data bound for, or already deployed, in the cloud, can be assessed and appropriately protected more rapidly, according to Scott Swalling, a data and cloud security expert at PA Consulting.

Swalling says cloud tools such as Google BigQuery and Amazon Macie use AI and ML to provide capabilities that help organisations better manage their data in public clouds and mitigate the exposure of sensitive data.

AWS Config, Azure Policy, or Google Clouds Security and Command Centre also help automate the monitoring and enforcement of security policies. Implementing continuous monitoring solutions will detect and alert on misconfigurations, suspect access requests and other security incidents in real time.

In addition to automated monitoring and enforcement, Swalling points out that the implementation of well-managed and regularly reviewed threat management allows organisations to be more proactive and agile in their response to threats.

Identity and access management is a core component of proactive IT security management. However, Carlos De Sola Caraballo, senior principal analyst at Gartner, warns that traditional asset-centric approaches to identity management will fail to provide the necessary visibility in cloud environments.

He recommends that IT security leaders focus on user identities and their associated permissions, establishing baselines for normal behaviour and configuring alerts to detect anomalies.

This approach enhances the ability to track and manage incidents across the cloud infrastructure, ensuring a more comprehensive and timely response, he says.

Whether an organisation is beginning its journey of migrating key services to the cloud or launching a cloud-native evergreen project, involving security specialists with a deep understanding of the cloud security model is an important factor.

Elliott Wilkes, chief technology officer (CTO) at Advanced Cyber Defence Systems, touches on the cloud shared responsibility model, whereby cloud providers are responsible for certain elements of each of the services. He says they need to monitor, defend and protect these elements, which include physical infrastructure and access controls at datacentres, resilient power backups and the like. All of the things youd typically expect a datacentre to provide, the CSPs [cloud service providers] will provide, he says.

Knowing what parts of the public cloud infrastructure are managed by the cloud service provider enables IT teams to develop a plan for how to tackle the security gaps they need to address.

Gartners Caraballo recommends that IT security leaders engage governance, risk and compliance (GRC) and legal teams early in the process of selecting a CSP.

Wilkes agrees, saying: Explicit contract stipulations are necessary to ensure robust incident response support from the CSP.

Caraballo recommends that IT security leaders consider overall business resilience when developing a strategy to respond to security incidents that occur in cloud environments. He notes that this requires a broader approach, which involves not only technical responses, but also strategic planning, such as digital supply chain redundancies and robust legal contracts. He urges IT security leaders to ensure their incident response plans are comprehensive, incorporating cloud-specific considerations and aligning with overall business continuity and disaster recovery strategies.

According to Caraballo, the transition to cloud environments necessitates a fundamental shift in incident response strategies. He urges IT security leaders to reassess and upgrade their incident response procedures, leveraging automation, proactive collaboration and identity-centric security to meet the unique challenges of the cloud.

The dynamic nature of cloud security demands equally dynamic and flexible incident response strategies, ensuring that organisations can respond swiftly and effectively to emerging threats, he adds.

The good news, at least from Swallings perspective, is that cloud providers have the ability to assess vast amounts of data and threats. This, he points out, means public cloud services are currently superior in leveraging AI than simpler on-premise security tooling.

See original here:
Security in the public cloud explained: A guide for IT and security admins - ComputerWeekly.com