ChatGPT爆重大漏洞：可能泄露1.2%用户数据 AI安全引人震惊,chat tools - 虚拟币交易所平台,数字货币,NFT
Exploding major vulnerabilities may reveal the security of user data, shocking the virtual currency exchange platform. digital currency officially released an investigation report on the temporary interruption of service on January, US time, and said that some users may have their data leaked. According to the official website announcement, the virtual currency exchange platform digital currency virtual currency exchange platform digital currency virtual currency exchange platform digital currency Home News Blockchain Blockchain Encyclopedia Home Encyclopedia Explosion major vulnerabilities may reveal the security of user data, which is shocking. Surprised, the U.S. official released an investigation report on the temporary interruption of service on May, and said that some users may have their data leaked. According to the official website announcement, an error in an open source library named before May caused a cache problem, which may cause users to see fragments of other people's chat records and even show the last four digits of other users' credit cards to some active users. The name, expiration date, e-mail address and payment address may have been sporadic users. However, the nerve-racking thing is that there was a mistake in making changes to the server on the morning of the th, which further aggravated the problem of requesting cancellation. Therefore, according to the official statement, there are two kinds of payment confirmation emails that may expose this information within a certain period of time, which may be sent to some wrong users. These emails contain the last four digits of the credit card. Clicking on my account management within a certain period of time may lead to the name of another active user. E-mail, payment address and the last four digits of the credit card were leaked, indicating that these users who may have leaked information have been contacted at present. In order to prevent such things from happening again, the actions they have taken were also explained in the blog post. Redundancy checks were added to ensure that the provided data matched the users, which improved the possibility of their clusters spitting out mistakes under high load, and emphasized that their open source maintainers have always been excellent collaborators, although the problems have been solved. There are still a large number of netizens who still question the security, and some netizens ask why they can't check the errors in the code. It may be to appease the user's mood. They immediately accepted an exclusive interview with the media on the same day. Although they didn't mention the incident of user data leakage, they emphasized security many times. He mentioned the following four points: the development and improvement of artificial intelligence language models need an iterative process including public feedback, focusing on solving prejudice and considering the progress of safety and consistency, especially technology. The development of artificial intelligence, especially neural network, may surpass human intelligence and greatly improve the quality of life, but it also brings great risks and challenges, which must be solved through continuous research and development of the security and consistency of artificial intelligence. It is a for-profit company that is committed to strong technology, while being transparent about security issues, and also issues its own views on artificial intelligence in official website. The system is becoming a part of daily life, and the key is to ensure that these machines conform to human intentions and values. Now, what they have created is challenging their commitment to the public. With the increasing frequency of using artificial intelligence in human work and life, security issues should also be a problem that we should pay attention to for a long time now and in the future. Technology should not only consider running fast, but also consider not overturning. According to the official explanation, this is due to the cache problem, which leads to the leakage of user information. However, this kind of mistake is not the first time in the internet industry. As early as Christmas in 2008, the personal information of about 10 users was leaked because of the cache problem. A huge amount of energy and money were invested in the research of potential security, but it was tripped over by a common old problem. Although it is not trivial to achieve great things, it is necessary to make great things easy in the world. I hope that this kind of problem will not happen again in the future. We asked how to solve the security problem and replied that it should be ensured in the field of artificial intelligence. The following are some suggestions and methods to solve security problems: interpretability and transparency. The design of the system should be interpretable so that researchers, users and regulatory agencies can understand its decision-making process. Improving transparency helps to ensure that the behavior of the system meets expectations and facilitate the discovery of potential security risks. Data security and privacy systems usually need a lot of data for training, and it is very important to protect the privacy and safety of these data by using technologies such as desensitization of encrypted data and differential privacy to ensure that. Data are protected in the process of processing, transmission and storage, and the system should have the ability to resist opponents, including identifying and resisting deceptive input of antagonistic attacks, studying robustness training methods, improving the anti-interference ability of the system, and establishing an auditable system to record the key decision-making process, which will help to track down potential problems and ensure compliance and provide a basis for correcting mistakes and misconduct in the future. Ethics and regulations should be followed to ensure that the system conforms to ethical principles and laws and regulations. Follow the principles of fairness, transparency and explanation, study ethics, formulate corresponding policies and guidelines, and continuously monitor and update the system with the progress of technology and the emergence of new security threats to ensure its safety and effectiveness. Multidisciplinary cooperation encourages researchers, policy makers, entrepreneurs and the public in different fields to participate in the discussion of security issues. Interdisciplinary cooperation will help to formulate a more comprehensive security strategy, open research and share, and carry out open and cooperative research to share security. Relevant knowledge, technology and best practices will help the whole field to make progress on security issues. By actively adopting these measures and methods, we can reduce the security risks in the field and promote the healthy and sustainable development of artificial intelligence. It is hoped that all artificial intelligence companies, including ourselves, will make more efforts to ensure security, so that the industry can develop more healthily. Author Lu Renjian typesetting Lu Renshang How to charge the back of the blockchain browser in official website's next decryption year? Post-business related articles cryptocurrency and five new trends worthy of attention in the field of cryptocurrency What does the optimism of cryptocurrency Ethereum blockchain mean? What are the characteristics of Ethereum blockchain? The allegations made by the Indian cryptocurrency exchange are false and unconfirmed. Seeking recourse to India's cryptocurrency policy, Bitcoin is still very popular. Five reasons why rich people choose to invest in Bitcoin? What are the hot articles? A brief introduction to the top ten bitcoin trading platforms? Bitcoin went on strike, but despite the price drop, investors remained calm. The latest trend of Bitcoin fell by half, and the price fell to a three-week low. 比特币今日价格行情网_okx交易所app_永续合约_比特币怎么买卖交易_虚拟币交易所平台
注册有任何问题请添加 微信：MVIP619 拉你进入群
打开微信扫一扫
添加客服
进入交流群
1.本站遵循行业规范，任何转载的稿件都会明确标注作者和来源；2.本站的原创文章，请转载时务必注明文章作者和来源，不尊重原创的行为我们将追究责任；3.作者投稿可能会经我们编辑修改或补充。
