What is the difference between Artificial Intelligence (AI) and Machine Learning (ML)? by Wong

Artificial Intelligence (AI) is the set of algorithms and techniques that allow a machine to perform tasks that are typically associated with human intelligence. AI also be a decision making and leads to the development of a system that stimulates human behavior in a given situation.     

​

Machine Learning (ML) is a subset of AI (Artificial Intelligence). ML only enables system to learn new skills from data and creating algorithms that learn on their own. Supervised Learning, Unsupervised Learning, and Reinforcement Learning are the three main methods of Machine Learning.

​

In simple words, AI need intelligence to solve complex problems which can perform by human while ML need to learn from data on specific task to maximize machine’s achievement in that task.

​

What are the fundamental punishments for a human and what practical punishments given that may achieve the same significance when imposed to a criminally liable AI entity? by Sim

(a) Death penalty 

• Considered the most severe punishment for humans 

• The most effective method of incapacitating offenders (since once the death sentence is carried out, the offender is obviously incapable of committing any further offense) 

• The significance of death penalty for humans: the deprivation of life 

• However, the “life” of AI entity is independent (may have physical appearance or only an abstract existence) 

• When an AI entity is justified to be guilty, the death penalty for AI is the deletion of the AI software controlling the AI entity as the deletion eradicates the independent existence of the AI entity 

• Thus, it has the same results as death penalty for humans 

 

(b) Imprisonment 

• One of the most popular sentences imposed in western legal systems for serious crimes 

• The significance of imprisonment for humans: the deprivation of human liberty & the imposition of severe limitations on human free behavior, freedom of movement and freedom to manage one’s personal life 

• In another way, the “liberty” of an AI entity implies the freedom to act as an AI entity in their relevant area 

• Therefore, the practical action that can achieve the same effect when imposed on AI entity is to put the AI entity out of use for a determinate period (no action relating to the AI’s freedom is allowed, thus its freedom is restricted) 

 

(c) Suspended Sentencing 

• A very popular intermediate sanction in western legal systems for increasing the deterrent effect on offenders replacing their actual imprisonment 

• The significance of a suspended sentence for humans: the threat of imprisonment if the human commits a specific offense or a type of specific offense 

• That person is imposed to a sentence of imprisonment for both offenses when they commit another offense during that period of suspended sentencing 

• The downside of suspended sentencing is that it is only imposed in legal records and no physical action is taken 

• Thus, there is no actual harm come to the AI entity if the relevant entity can be identified specifically and accurately 

 

(d) Community Services 

• Another popular intermediate sanction in western legal systems substituting the actual imprisonment 

• The significance of community services for humans: compulsory contribution of labor to the community 

• Most of the AI entities work for the benefit of the factory owners or other workers to ease and facilitate their professional tasks 

• Therefore, AI entities can also work for the benefit of the community when AI entities are sentenced to community services for their offences 

• In the end, the significance for both humans and AI entities are identical 

 

(e) Fines 

• The most popular intermediate sanction in western legal systems in lieu of actual imprisonment 

• The significance of paying fines for humans: the deprivation of some of their property (money/ property) 

• The penalty of paying fines by a corporation have the same meaning as paying fines by an individual (both have property and bank accounts) 

• However, most AI entities do not have its own property or money 

• Therefore, when a fine is imposed on an AI entity, it might be collected in the form of labor for the benefit of the community and some as property or money if that AI entity have properties or money   

 

In conclusion, most punishment imposed on humans have the same significance when imposed to a criminally liable AI entity. Some general punishment adjustments are necessary so it will produce the biggest effect to the AI entity. However, whether the AI entity will commit the same offense again or have any reflection upon their mistakes will be another question in another day.

​

Can AI take over the world? by LILIAN

Science fiction (Sci-fi) movies have changed some people’s perception of Artificial Intelligence. They image robots programmed with AI will become so powerful that they will eventually destroy their creator and destroy the entire world leading to the development of a new AI-controlled species.An intriguing storyline, but untrue for it to realistically occur anytime in the near future!
 

It is undeniable that Artificial Intelligence has come a long way and developed into a unique feature of the modern world. Despite the advancements in AI, most of the tasks are still done under some kind of human supervision in the working or the development stages.
 

Artificial Intelligence is also limited to be the particular task that it is programmed to complete. An interesting example is self-driving cars, where AI takes control of the car and drives it to the desired destination of the user’s choice. However, the AI is only limited to the driving of the car precisely and no other external tasks.
 

So, for AI take over the world as shown in the sci-fi movies, we are still at least a few decades away from such an outcome. But, just to keep the curiosity around this speculation, it could eventually be possible in the future! Although currently, this is only fiction.

​

Is AI smarter than humans? by Kelly

AI is remarkably useful and can solve complex problems that humans cannot solve. It is also faster on the right tasks. In some cases, AI can determine better results than human-based decision matrices. This is based on its ability to recognize complex patterns in large amounts of data. However, the ability of artificial intelligence to independently execute complex divergent thinking is extremely limited. In other words, artificial intelligence is not smarter than humans.