The benefits along with risks of artificial intelligence
Self- driven cars to automated machines the concept of artificial intelligence is on a fast lane. The short form of artificial intelligence is AI which might undertake a narrow task but the long term implications are multitude. AI is expected to outnumber human beings in a variety of tasks.
Is research of AI in terms of safety important?
In the days to come the impact of artificial intelligence service companies would mitigate research in various areas of behavioural studies. In fact it could be a minor form of nuisance if your laptop goes on to crash but it would be really important to understand that your AI performs what you want them to do. In the future a major point of debate might centre on the fact what happens if AI system replaces human intelligence.
Could you term AI to be dangerous?
Researches in most artificial intelligence service companies agree that AI is not likely to toy with human emotions or feelings. No reason to expect that AI is going to become benevolent. Now considering the reason why AI could pose to be a risk there are two pointers to consider
- Programming of AI could take place in a manner that can lead to something disastrous. – The autonomous weapons are AI modules which could programme to kill. If the weapon lands in the hand of a wrong person it can lead to causalities. At the same time an AI arm race could lead to an AI war leading to a lot of causalities. When you are attacked by enemies such weapons would be really difficult to turn off. This risk could be posed with a narrow AI but it is expected to grow once the level of AI increases with autonomy increasing
- The programming of AI could be developed for something positive but negative events could appear- this could arise due to a situation where it becomes difficult to align AI goals with ours. In case if you ask an obedient car to take you over to a hospital you might reach there but might be chased by helicopters. Ideally this is a situation that you do not want to land up.
All these aspects can lead to incompetence of AI systems. A superior AI system should help you make tasks easier and they need to align with your goals. If this is not the case then there poses to be a definite problem.
Why AI safety has become a topic of debate?
Lot of popular names of late have expressed their concern about AI safety which is supported by AI researches. Now the question is why is the subject all of a sudden in the headlines.
As AI is expected to replace human intelligence, there is no clear cut way to predict on how it is going to behave. The best example in this regard would be our own evolution. People not control the planet, not because we are the strongest, but as we are the smartest.