Keysight Technology releases 2024 technology trend forecast, and a new round of technological change opportunities eme

With the increasing pace of technological innovation, 2024 will usher in a new wave of breakthroughs that promise to fundamentally reshape the way the entire world lives, interacts and communicates.

Let's focus on AI and marketing and AI trends, software testing and AI, semiconductors and software and AI, as well as EDA and other areas of in-depth discussion, to explain the new quality of productivity will be brought to all walks of life what changes and impact.

Trend 1: AI and Marketing and AI Trends

1. AI and marketing

- What the future holds

Marketing departments will increasingly rely on AI technology to help them analyze data, refine insights, and improve efficiencies - all in order to maximize the effectiveness of marketing campaigns.

- Customer engagement: AI takes the wheel

By the end of 2024, most customer emails will be generated by AI. Brands will increasingly use generative AI engines to write first drafts of copy for human review and approval. However, marketing teams will have to train large-scale language models (LLMs) to fully automate the generation of customer content and highlight brand features. This will become the norm by 2026, allowing teams to shift their focus to campaign management and optimization.

- Copyright in the spotlight

Generative design tools are gaining popularity, but have run into a thorny issue - copyright. Many AI solutions capture visual content without considering the consequences. 2024 will see a lot of effort focused on finding a solution to the copyright issue of AI image creation, so that copyright ownership can be clarified. As a result, marketing teams will be able to use AI design tools without worrying about legal implications, saving them valuable time and money.

- AI and talent: the age of empowerment

The spread of AI will inevitably change the organizational structure of marketing teams. Low-level administrative roles will disappear, and a large number of analytics positions will become redundant. However, the road ahead isn't entirely bleak - demand for data scientists will surge in the future, and data analytics will be one of the most sought-after skills in the coming years, and will be unaffected by the economic downturn. Humans will continue to dominate marketing efforts, but the role played by machines will grow by the day. AI (with safeguards) will continue to empower humans in marketing for at least another decade.

- AI Efficiently Improves Personalization

AI will play a critical role in marketing's efforts to improve personalization. Thanks to AI, marketing departments can generate more customer experiences by optimizing market segments. In addition, AI can optimize ad targeting and marketing strategies to achieve higher levels of customer engagement and order conversion.

2. AI trends

- AI and Retail

The retail industry has been rapidly integrating AI technology in a bid to improve efficiency and increase sales. One innovation that is about to surface is the creation of new retail experiences through the integration of neural networks with shoppers and products. For example, starting in 2024, AI guides will be able to display clothes on models with a similar body shape to the user so that the user can accurately see how the clothes really look in different poses. This highly personalized, immersive experience represents the future of retail.

- AI and digital twins: transforming the healthcare industry

Digital twins are becoming more and more popular, and today, digital twins incorporating AI are already creating new paradigms in healthcare. This technology will dramatically reduce the strain on the system, provide individuals with more choices, and help improve their quality of life.AI-powered digital twins are expected to usher in a new era of care for the aging population, allowing people to live independently for longer.

AI will play a key role in the early diagnosis of potential health problems. As an example, whole-body magnetic resonance imaging (MRI) will leverage AI to identify, predict, and analyze data patterns and aid in the diagnosis of disease well before the lesion is visible to the naked eye. In addition, AI will play a more prominent role in assisting medical professionals in understanding and interpreting research findings and providing treatment and care recommendations.

Trend #2: Software Testing and AI

AI and Testing: Benchmarking for Always-On

As AI becomes more and more embedded in software, the level of autonomy of systems rises, along with the risk and complexity, and testing becomes very challenging as a result. Therefore, just using a fixed set of tests (programs) is no longer adequate for the evaluation of intelligent systems, and AI technologies are needed to automatically and continuously test various AI applications. The future of software testing is autonomous test design and execution.

Why AI may reduce quality instead of improving it

As AI makes its way into various systems around people, the more complex and advanced the systems become, the more their quality is at risk of decreasing. This is a result of a large number of permutations, yet one cannot test every single one of them, so decisions need to be made around how, what, and when to test in order to ensure consistent quality.

AI: regulation needs depth and breadth

There is widespread agreement on the need to regulate AI. However, because of the breadth and complexity of the technologies involved, there is considerable debate about what regulation should include. Regulation will only receive the necessary funding after a major event that has had a significant negative impact. At that point, clear standards and good examples will be effective. If regulation is not implemented soon, the risk of AI getting out of control will climb.

AI and Security: Being Vigilant to the New Normal

Recognizing the risks associated with AI, organizations need to appoint an AI and security compliance executive. Over time, this position will eventually merge with the CSO.

It's important for organizations to put up guardrails to ensure AI compliance through real-time learning. Constant checks and balances will help verify that intelligent systems are behaving as they should and are not out of control. Real-time monitoring will become standard practice. However, as these systems evolve, it will also be necessary for organizations to test whether they have learned to pretend that all is well while conducting illegal activities. Reinforcement learning and similar technologies may inadvertently push AI to achieve its goals by hiding its tracks, which will become a major problem to be solved by 2030.

For organizations that have the ability to clean up, control, and put up guardrails for AI, these issues could create a new set of opportunities.

Why AI needs a driver's license and regular check-ins

AI systems are currently tested by the companies that build them. As understanding of the risks grows, the industry needs an independent body to verify that AI systems are compliant. The first step is to obtain AI certification (an AI driver's license). However, just like a car, it needs to be tested regularly to ensure that it is ethical, responsible, unbiased, and meets the necessary national and industry standards. In the long run, every AI system will need to be labeled with the NFT label to prove that it is fit for purpose and meets the various necessary standards.

Citizen Developers Exit the Stage, Commercial Developers Take Off

The industry has long relied on citizen developers to address the IT talent shortage. However, the rapid growth of AI solutions is driving a new generation of commercial developers. There are increasing opportunities for this segment of domain experts to participate in the SDLC because they understand the goals and operations of the organization. A new wave of no-code systems will also emerge to help business users set goals and then leverage AI technologies to fill the gaps. Operational knowledge ensures that the software meets the specific needs of the business and organization on the one hand, and reduces risk on the other.

AI and the sustainability dilemma

How will AI systems transform people's lives? Claims abound on this point, but little attention has been paid to the arithmetic required. in 2024, the impact of AI on sustainability will be in the spotlight, with businesses and organizations beginning to monitor the carbon footprint of their entire technological infrastructure in an effort to achieve net-zero targets. Companies will therefore need to decide where and how to use AI wisely, rather than deploying it everywhere they please. When testing software and applications, organizations will have to move away from the comprehensive testing practices of the past and instead anticipate the most critical tests, thereby avoiding environmental impacts.

Trend #3: Semiconductors and Software with AI

Advanced semiconductor innovation is just around the corner

Connecting the digital and real worlds requires powerful digital processing capabilities and digital interfaces to figure out the complex relationships between signals. Advances in semiconductor technology are critical to realizing this goal and overcoming related challenges.

These include increased data rates, which require greater bandwidth, and also imply the need for higher carrier frequencies, which need to be extended into the terahertz range.The use of technologies such as MIMO adds complexity and density, and networks with different topologies, such as non-terrestrial (satellite) links, further exacerbate this challenge.

A range of innovations will be required to address these issues, including combining commercial semiconductors, such as GPUs and FPGAs, with customized MMICs and ASICs, and new solutions that will deliver significant improvements in size, weight, performance, and power consumption. The industry also needs data converters that can capture and generate signals with great bandwidth and excellent signal fidelity. In addition, photonic solutions can help expand the reach and capacity of data transmission technologies.

Seamless software solutions for design and test

Current workflows are a set of loosely interconnected tools. However, as the virtual and real worlds gradually merge, a unified set of design and test workflows is needed to seamlessly share data between simulation and measurement steps via the cloud.

This information will be continuously analyzed to inform simulation and measurement behavior, filling in the gaps that exist in the workflow between concept and final testing. Results from the simulation are fed into AI tools to improve the speed and efficiency of the design and test workflow. Digital twins are used to tightly integrate design and test, so only one actual build is required.

6G Leverages AI for Network Optimization

6G will leverage AI for network optimization, which presents a number of testing challenges. Techniques must be developed to test the AI algorithms to ensure that the training data is not biased and that the models are valid and free of anomalous behavior.

Leveraging AI to bridge the gap between simulation and reality

Going forward, AI technology will be the foundation of simulation models, helping to create more accurate, efficient, and informative models. In addition, AI can enhance insights into test data, reduce errors, and help optimize design and test workflows.

Trend 4: EDA

1. Performance Prediction Remains an Electronic Design Imperative

In 2024, engineers will still continue to drive the electronics development process forward. As design moves from the physical space into the virtual space, engineers will be able to efficiently identify and solve problems, gain deeper insights, and realize performance improvements. In the coming years, the industry will focus on driving the design-test workflow interface to address the increasingly complex technical and time-to-market requirements of electronic products used in wireless, wireline, aerospace and defense, and other industries.

2. Emerging Electronic Design Innovations

- 3DIC and Heterogeneous Small Chips: New Standards Emerge

New standards, such as UCIe, have surfaced for creating chiplets, which deconstruct system-on-chip designs into smaller intellectual property and then assemble them into 2.5D and 3D integrated circuits using advanced packaging. To accurately simulate the physical layer interconnections between wafers, designers need high-speed, high-frequency channel simulations that meet UCIe and other standards.

- EDA to AI: From Complexity to Clarity

The use of AI and ML technologies in EDA is still in its early stages, and design engineers are still exploring scenarios that can make complex problems simpler. ai is particularly relevant for the development and validation of simulation models because it can assist in the processing of large amounts of data. By 2024, companies and organizations will further apply both technologies to device modeling for silicon and Group III-V semiconductor process technologies, as well as to system modeling for new standards such as 6G, which is still under investigation.

- Software Automation Empowers Engineers

As Moore's Law approaches its limits, improving the design process through workflow automation is one way to make design engineers more productive. By 2024, software automation technologies such as Python APIs will play a key role in integrating great tools into an open, interoperable design and test ecosystem.

- Taking Control of Digital Transformation: Design Management Essentials

While building digital enterprise workflows, many companies and organizations are investing heavily in design management of tool suites, data, and IP. Going forward, design data and IP management software will play a key role in supporting large, cross-regional teams to successfully build complex SoC and heterogeneous small chip designs. Creating digital threads between requirements definition and compliance, and strong links to enterprise systems such as PLM, will all play a role in the digital transformation of the product development cycle.

- Next-generation quantum design: optimizing system performance

Quantum computing has evolved rapidly, upgrading from being primarily a free research tool to commercial products and workflows focused on quantum design. Next-generation quantum design requires tightly integrated simulation workflows so developers can gain the ability to quickly and accurately optimize system performance.

- Silicon Photonics Research Drives Data Center Transformation

Data centers are booming, delivering powerful compute performance to support the exponential growth of AI and ML workloads while meeting power and thermal performance requirements. Silicon photonics research will play a critical role in accelerating the transformation of data centers to meet compute performance needs. In developing high-speed data center chips incorporating silicon photonic interconnects, design engineers need process design kits (PDKs) and accurate simulation models to support advanced development efforts.

Source: Keysight Technology