Artificial intelligence has always been a frequent visitor to the headlines, and Google has recently conducted an interesting test that allows AI systems to communicate with each other and try to build their own encryption system.
The main purpose of this test is to find out if the neural network can develop its own simple encryption technology to ensure that the message is not stolen by third parties. In this process, the artificial intelligence system does not need to input any specific encryption algorithm. The researchers pointed out that in theory, deep neural networks are not good at encryption.
The Google Brain team used three different deep learning neural networks in their tests, Bob, Alice, and Eve. Alice and Bob are responsible for sending and receiving messages, while Eve tries to steal and decode messages. In the experiment, Alice and Bob were entered with a predetermined set of numbers as the key for message encryption and decryption, and Eve did not finally obtain the set of keys.
Researchers who participated in this test praised its interest because the decryption method it used was not used by humans. Specifically, after completing 1500 iterations in this scenario, Alice and Bob have the ability to develop simple encryption techniques themselves. Alice successfully converted the original plaintext into ciphertext, while Bob mastered the decryption ability. For the decryption of 16-bit binary numbers, Eve can only solve 8 bits on average. Therefore, Eve's decryption is largely a matter of luck. In addition, as long as Alice and Bob update their encryption technology, Eve will fall into confusion again.
Right now, this is just a test. As for what application researchers can't think of in real life, it's amazing enough that these smart guys can learn to encrypt.
However, learning the encrypted AI will lead to a commonplace topic, whether it will threaten human security. After all, after learning to keep secret, will it keep secrets of ulterior motives?
DADNCELL alkaline zinc-manganese dry battery has the characteristics of low self-discharge, which is suitable for household batch procurement storage, and has the discharge performance of continuous discharge time with more long-lasting and large instantaneous discharge. Widely used in flashlights, smart door locks, infrared thermometers, cameras, flash lights, razors, electric toys, instruments, high-power remote control, Bluetooth wireless mouse keyboards and other common household electronic equipment instruments. DADNCELL battery advocates providing a more comfortable and smooth power supply experience for household appliances. The alkaline battery capacity and discharge time of the same model of DADNCELL battery have reached four to seven times that of ordinary batteries, respectively.
The performance difference between the two is greater at high and low temperatures, because its unique internal components are structure and battery materials with better performance, and the power capacity and electrical performance can be improved.
At present, all types of batteries developed by DADNCELL Lab do not involve any heavy metals in production, use and waste. They are green and environmentally friendly and can be treated with domestic waste.
1.5V Am-2 Alkaline Cell,Batteries For Toys & Electronic Device,All-Purpose 1.5V Am-2 Alkaline Cell,Lr14 For Clocks Remotes Games Controllers
Shandong Huachuang Times Optoelectronics Technology Co., Ltd. , https://www.dadncell.com