Opinion on AI risk management

Vision-based AI detection technology can identify and analyze image content through deep learning algorithms, and quickly and accurately identify illegal, harmful or false content, such as pornography, violence, authenticated materials, etc.

There are also some risks and challenges when children use AI, such as privacy protection, information security and other issues. Therefore, parents and educational institutions need to guide and supervise children’s correct use of artificial intelligence technology to ensure that they can benefit from it without being harmed.

Human-made abnormal operations have caused AI to become a tool for crime. In China, AI face-changing technology has become one of the operations of new financial fraud crimes.

Other country crime cases:

1. In 2016, a programmer was arrested in Russia because he developed a malware called Mirai that was used to launch large-scale DDoS attacks that disrupted Internet services worldwide.

 2. In 2018, an American high school student used AI technology to forge the voice of a company’s CEO, and requested the financial department to transfer large sums of money to his designated account over the phone, causing financial losses to the company.

 3. In 2019, a hacker used AI technology to forge the voice of a company’s employees and send false information to customers over the phone, causing customer information to be leaked and the company’s reputation to be damaged.

 4. In 2020, a hacker used AI technology to generate a large number of fake social media accounts and used these accounts to spread false information and malicious links, causing confusion in public opinion and network security issues.

 5. In 2021, a hacker used AI technology to attack the database of a medical institution, stole a large amount of patients’ personal health information, and demanded a ransom to prevent the information from being leaked.

AI risk management lies in the accuracy of system assessment, the vastness of technology cloud algorithms, and a series of structural actions.

The best way to manage AI risks is with the intervention of governments and technology companies. AI can achieve real-time monitoring and rapid response to prevent the spread of objectionable content.


Comments

Leave a Reply

Your email address will not be published.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