DeepNude Application
Recently, an American programmer developed an application called DeepNude, which became popular across various platforms. Currently, the DeepNude official website has ceased its online functionality but offers software downloads, and some foreign users have released cracked versions (this account, historical messages, daily updates, links), usable on Windows 7 (64-bit) and above. You may not understand DeepNude (and it’s probably best not to), but as long as you grasp the concept of artificial intelligence, that’s sufficient. Its principle is based on the open-source algorithm pix2pix developed by researchers at the University of California, Berkeley. This software utilizes algorithms similar to those used in autonomous vehicle technology, training the computer with a database of 10,000 images.
Artificial Intelligence Mimics Its Creator
Artificial intelligence is a branch of computer science that attempts to understand the essence of intelligence and produce a new type of intelligent machine that can respond similarly to human intelligence. Simply put, it involves injecting values into computers and teaching them to think like humans (such as learning, reasoning, thinking, planning, etc.), enabling them to make judgments and react independently without human intervention. Here’s an example from the field of autonomous vehicles that illustrates the thought process of artificial intelligence.
First Principle of Autonomous Vehicle Thinking:
Do not harm humans, even at the cost of self-destruction.
Second Principle of Autonomous Vehicle Thinking:
In unavoidable collision scenarios, protect the driver’s safety.
At first glance, this thought process seems flawless, but how should an autonomous vehicle operate when it must sacrifice itself to ensure pedestrian safety? A typical human driver would choose self-destruction, a virtue inherent to humanity. But can a computer make such a choice? This poses a profound philosophical question.
In some respects, artificial intelligence is equivalent to the designer’s thought patterns and worldview; it follows the standards set for it.
Who Does China’s Artificial Intelligence Resemble?
Didi – The “Little Knife” in Artificial Intelligence
Didi, developed by Beijing Xiaoju Technology, is also a major brand in China. After enduring numerous challenges and burning through vast amounts of money, it has finally unified the taxi industry, and now it’s time to collect fees. How to charge? By using artificial intelligence to exploit user familiarity! Didi’s starting point is the same, but different fares for different users is a result confirmed by many. Didi can determine a user’s sensitivity to price and distance based on trip data and ride-hailing data from the app. For example, if you frequently pay extra for rides, the AI will interpret this as a lack of price awareness. Similarly, if you take the same route during a specific time period, the AI will assess that the time cost of this journey is high, making your fare more expensive than others. Each person pays a little more, Didi has truly taken Jack Ma’s philosophy to the extreme. The solution is to log in with a different number or call rides for each other.
Baidu – The “Stirring Stick” of Artificial Intelligence
Baidu’s brand has proven to be remarkably resilient in the IT sector, especially as many of its contemporaries have faded into obscurity. It often seems on the verge of collapse, yet it somehow manages to survive, still holding the title of China’s top search engine, although its effectiveness is questionable. From Baidu Doctor a few years ago to the current Baidu Cloud, the company has exhibited issues of character. If you use Baidu’s search engine, it automatically ranks search results based on the choices of “the majority”, regardless of quality, accuracy, or ethical boundaries. This frustrating algorithm can be quite exasperating. Baidu may cater to the general public’s taste, but it is not suitable for everyone. Baidu knows, but you know nothing! The solution is to use specialized search engines or navigational tools for mutual supplementation.
ByteDance – The “Tamer” of Human Thought
No specific product name is mentioned because ByteDance has too many products, such as the famous Toutiao and Xigua Video, among others. ByteDance is a company with immense content creation and intelligent innovation capabilities in China. However, as an AI enterprise, their mission is not to tame artificial intelligence but to use it to train human thought. Their primary method is “intelligent push”. Similar to Didi, user click data from mobile devices undergoes intelligent analysis to summarize user preferences and habits, leading to AI algorithm training. Based on these algorithms, personalized information is then pushed to users, gradually narrowing the push range until it locks onto their thinking. At this point, users will find the application incredibly useful, filled with content they like. In reality, users have been mentally “farmed” by AI, losing their ability to explore the unknown world through apps and becoming mere tools for increasing click rates. The solution is to frequently switch between similar apps to prevent being locked in by artificial intelligence.
There are many similar companies in China, such as Taobao, but having mentioned Didi, there’s no need to discuss Taobao further. We cannot always criticize Jack Ma, as Alibaba’s artificial intelligence ranks second in the world, and we should still rely on Alibaba to bring glory to the country.
DeepNude Application and Chinese Artificial Intelligence
Although the final result of the trained DeepNude is “tragically unwatchable” (very unsuccessful, with negligible effects), many people criticize DeepNude excessively, which is unnecessary. Compared to Chinese artificial intelligence companies, although the nature is different, when viewed through certain standards (where discussing ethical boundaries is inappropriate), they are not much different in essence.
In the future, artificial intelligence will be the “container” of human wisdom. What exactly are these AI companies doing to computers? Based on the current state of these companies, if one day artificial intelligence can fully simulate human consciousness and thought, it will inevitably lead to a disaster for humanity.
The First Lesson AI Should Learn is “Moral Thinking”
Follow the QR code for public account historical messages, daily updates
Push new benefits
