What can edge artificial intelligence (Edge AI) do for us?

• Artificial intelligence (AI) is a type of decentralized computing that allows devices to make data-driven decisions at the closest point of interaction with the user.
• The benefits of this technology include improved privacy and cost savings, but the data is typically discarded after processing.
• Coming advances, including 5G technology and lower-cost processing chips, will make edge AI increasingly useful for certain applications — from smart home devices to medical technology.

Imagine that you want your new smart thermostat to quickly turn up the temperature so your house is warm when you get home from get off work on an unusually cold day. You connect from your smartphone and ask it to take action. You wouldn't know it, but the operation may take a few seconds as it sends your request to the cloud and receives the instructions back.

Now imagine that the self-driving car you are riding in suddenly senses a dog running onto the road in front of you. Cars need to react within milliseconds to avoid disaster. This response requires edge artificial intelligence (AI) – technology that can make decisions at the closest point of interaction with the user, and in this case, the car’s sensors are crucial. This is the definition of a split second decision.

Dynamic Data

With today’s Internet of Things (IoT), data is always in motion. It flows from legacy systems to the cloud, all the way to edge devices, and beyond an organization’s systems to partners and customers. Answers need to be delivered in real time, so using centralized computing power is not always efficient when the data can be processed via edge devices. When a self-driving car only has milliseconds to react, it doesn’t have time to wait for the cloud to make a decision.

No matter where the device is located, vast amounts of data can be fed into AI algorithms at the edge, and the benefits are numerous. Dynamic data can deliver important patient information to doctors, shorten queues at amusement parks, alert power companies to potential power outages, and enable self-driving cars to respond in time to prevent tragedy.

Edge AI allows devices to make these decisions on their own at the device level. It does not necessarily have to be connected to the Internet to process data. Consider a watch that monitors your sleep patterns, but instead of pushing the data to the cloud for storage and processing, it records the data on the watch itself for processing.

Edge-enabled AI devices also include video games, smart speakers, drones and robots. Security cameras can also enable edge capabilities – cameras on the factory floor look for product defects during the manufacturing process and can quickly identify which products need to be pulled out immediately. When speed saves lives, edge AI can also be used to analyze images for emergency medical care. The closer the processing capabilities are, the faster the response time.

While edge technology won’t replace the cloud, user data that belongs only to you (such as your sleep patterns or gaming data) can be processed in edge-enabled devices. This decentralization of data solves privacy concerns, which is an important issue in the IoT market. Edge

AI can provide convenience without compromising privacy. And, in some cases, it could be cheaper—one company is currently developing voice-controlled home appliances, such as washing machines and dishwashers, using tiny microprocessors that cost a few dollars each.

“When it comes to the gadgets in my house, I actually wish they were less smart.” — Clive Thompson, Wired

For example, the speech recognition AI of a coffee machine only needs to recognize about 200 words, all of which are related to the task of making coffee. Think about it, says Wired reporter Clive Thompson: "I don't need a bad joke or a light switch to achieve self-awareness. They just need to recognize "on" and "off" and maybe "dark." When it comes to the gadgets that share my house, I'd actually prefer them to be less intelligent."

In addition to faster and cheaper processing, edge AI doesn't require the ever-expanding internet. With the rapid growth of the Internet of Things, vast amounts of data are now being sensed and generated at the edge—Statista estimates that this number will reach nearly 80 zettabytes by 2025.

This is so huge that using the bandwidth of today’s Internet to transfer all this data from edge devices to cloud servers for storage and processing is technically infeasible. Even if bandwidth is available, there needs to be enough data center resources to handle all the data. Less bandwidth requirements translate into cost savings. About 10% of enterprise-generated data is created and processed outside of traditional centralized data centers or the cloud. Gartner predicts that this number will reach 75% by 2025.

Balancing Risks and Rewards

One of the most vexing issues in the IoT world is that the large number of people who cannot afford devices or live in rural areas without local networks may not be able to participate in the impact on our daily lives transformation. A history of limited network capacity can become a vicious cycle. Edge networks are not simple to build and can be expensive. Developing countries may fall further behind in their ability to process data via edge devices that require newer technologies. The growth of edge computing is therefore another way in which structural inequality may increase, particularly as it relates to the accessibility of life-changing artificial intelligence and IoT devices.

Another risk with edge AI is that data may be discarded after processing—by its very nature “at the edge,” this means it may not make it to the cloud for storage. Devices can be instructed to discard information to save costs. While there are certainly disadvantages to central processing and storage, the advantage is that the data is there when you need it.

Artificial Intelligence, Machine Learning, Technology

How does the World Economic Forum ensure that the development of artificial intelligence benefits all stakeholders?

Artificial intelligence (AI) is impacting every aspect of society—homes, businesses, schools and even public spaces. But as technology rapidly evolves, multi-stakeholder collaboration is needed to optimize accountability, transparency, privacy, and fairness.

World Economic Forum’s platform for shaping the future of technology governance: Artificial intelligence and machine learning are bringing together diverse perspectives to drive innovation and build trust.

• One of the areas of work that can make the most of artificial intelligence is human resources—including recruiting, retention, training, benefits, and employee satisfaction. The Forum created a Human-Centered HR AI Toolkit to promote positive and ethical human-centered use of AI by organizations, workers and society.
• Today’s children and young people are growing up in an increasingly digital age, where technology permeates every aspect of their lives. From robotic toys and social media to classrooms and homes, artificial intelligence is a part of life. By developing AI standards for children, the Forum is working with a range of stakeholders to develop actionable guidelines to educate, empower and protect children and young people in the age of AI.
• The potential dangers of artificial intelligence may also affect wider society. To reduce risks, the Forum brings together more than 100 companies, governments, civil society organizations and academic institutions in the Global AI Action Alliance to accelerate the adoption of responsible AI in the global public interest.
• Artificial intelligence is one of the most important technologies in business. To ensure C-Suite executives understand the possibilities and risks, the Forum created Empowering AI Leadership: AI C-Suite Toolkit to provide practical tools to help them understand the impact of AI on their roles and make decisions on AI strategy, projects and implementation Informed Decisions.
• Shaping the way artificial intelligence is integrated into public sector procurement processes will help identify best practices that can be applied across the private sector. The Forum has developed a set of recommendations designed to encourage widespread adoption, which will evolve with insights from a series of trials.

If it’s just you and your self-driving car driving on an empty road, that large amount of data may not seem important, but think again. A lot can be learned from the data on this empty road, including information about road conditions and how the vehicle and others like it behave under those conditions. Finally, when it comes to edge computing, a clear business case must be scrutinized to ensure the cost of the network is balanced against the value created.

Still, despite the inequality or data loss, and with advances in 5G technology and cheaper processing chips, it’s easy to see how “at the edge” is here to stay — whether that’s you The self-driving car is also your coffee maker, getting you ready for your commute.

The above is the detailed content of What can edge artificial intelligence (Edge AI) do for us?. For more information, please follow other related articles on the PHP Chinese website!