Hand touching a node in an AI edge network represented by a brain.

The risks of advanced AI

AI at the edge can revolutionize your business, but what do you need to avoid unintended consequences?

Hand touching a node in an AI edge network represented by a brain.
Image: stnazkul/Adobe Stock

With the growing demand for faster results and real-time insights, companies are turning to cutting-edge artificial intelligence. Edge AI is a type of AI that uses data collected from sensors and devices at the edge of a network to provide actionable insights in near real-time. Although this technology offers many advantages, its use also carries risks.

SEE: Don’t Curb Your Excitement: Trends and Challenges in Edge Computing (TechRepublic)

Jump to:

Cutting edge AI use cases

There are many potential use cases for artificial intelligence at the edge. Some possible applications include:

  1. Autonomous vehicles: Edge AI processes data collected by sensors in real time to decide when and how to brake or accelerate.
  2. Smart factories: Edge AI monitors industrial machinery in real time to detect anomalies or faults. The cameras also detect defects on the production line.
  3. Health care: Wearable devices can detect heart irregularities or monitor patients after surgery.
  4. Detail: Store sensors that track customer movement and behavior.
  5. Video analysis: AI analyzes video footage in real time to identify potential security threats.
  6. Facial recognition: Edge AI can be used to identify individuals by their facial features.
  7. Speech Recognition: Edge AI is now used to recognize and transcribe spoken words in real time.
  8. Sensor data processing: Edge AI can process data collected by sensors to decide when and how to brake or accelerate.

Edge AI Risks

Data lost/rejected

Edge AI risks include data that may be lost or deleted after processing. One of the benefits of advanced AI is that systems can delete data after processing, which saves money. The AI ​​determines that the data is no longer useful and deletes it.

The problem with this configuration is that the data is not necessarily useless. For example, an autonomous vehicle can drive down a deserted road in the remote countryside. The AI ​​can deem most of the collected information useless and discard it.

However, data from an empty road in an outlying area may be useful depending on who you ask. Additionally, the collected data may contain information that may be useful if it reaches the cloud data center for storage and further analysis. This could, for example, reveal animal migration patterns or changes in the environment that would otherwise go unnoticed.

An increase in social inequalities

Another risk associated with AI is that it can exacerbate social inequalities. This is because Edge AI needs data to work. The problem is that not everyone has access to the same data.

For example, if you want to use Edge AI for facial recognition, you need a database of face photos. If the only source of this data comes from social networks, the only people who will be accurately recognized are those who are active on social networks. This creates a two-tier system in which cutting-edge AI accurately recognizes some people while others don’t.

SEE: Artificial Intelligence Ethics Policy (TechRepublic Premium)

Additionally, only certain groups have access to devices with sensors or processors capable of collecting and transmitting data for processing by state-of-the-art artificial intelligence algorithms. This could lead to a situation where social inequalities increase: those who cannot afford the devices or who live in rural areas where local networks do not exist will be excluded from the AI ​​revolution of point. A vicious cycle could result, as edge networks are not simple to build and can be expensive, which means the digital divide can increase and disadvantaged communities, regions and countries can fall further behind in their ability to leverage the benefits of cutting-edge AI.

Poor data quality

If the sensor data is of poor quality, the results generated by a state-of-the-art artificial intelligence algorithm may also be of poor quality. This could lead to false positives or negatives, which could have disastrous consequences. For example, if a security camera using state-of-the-art AI to identify potential threats produces a false positive, it could lead to innocent people being detained or questioned.

On the other hand, if the data is of poor quality due to poorly maintained sensors, it could lead to missed opportunities. For example, if an autonomous vehicle is equipped with advanced artificial intelligence used to process sensor data to make decisions about when and how to brake or accelerate, poor quality data could drive the vehicle making bad decisions that could lead to an accident.

Poor accuracy due to limited computing power

In typical edge computing configurations, edge devices are not as powerful as the servers in the data center to which they are connected. This limited computing power can lead to less efficient state-of-the-art artificial intelligence algorithms, as they must run on smaller devices with less memory and processing power.

Security vulnerabilities

Edge AI applications are subject to various security threats, such as data privacy disclosure, adversarial attacks, and privacy attacks.

One of the biggest risks of cutting-edge AI is data privacy disclosure. Edge clouds store and process large amounts of data, including sensitive personal data, making them attractive targets for attackers.

Adversarial attacks are another risk inherent in cutting-edge artificial intelligence. In this attack, an attacker disrupts the input of an AI system to cause the system to make an incorrect decision or produce a false result. This can have serious consequences, such as causing a self-driving car to crash.

Finally, AI systems at the edge are also vulnerable to privacy or inference attacks. In this attack, an attacker attempts to discover the details of the algorithm and reverse engineer it. Once the correct inference is made about the training data or the algorithm, the attacker can make predictions about future inputs. Edge AI systems are also vulnerable to various other risks, such as viruses and malware, insider threats, and denial of service attacks.

Balancing risk and reward

Edge AI has benefits and risks; however, you can mitigate these risks with careful planning and implementation. When deciding whether or not to use cutting-edge AI in your business, you need to weigh the potential benefits against the threats to determine what is right for your specific needs and goals.

#risks #advanced

Leave a Comment

Your email address will not be published. Required fields are marked *