The Reality of AI | Pt.7 | AI is a programme. A programme is only as reliable as the programmer.

Even if AI reaches the stage where it appears to be fully sentient and able to think fully for itself, the truth that everyone will sooner or later have to accept is that AI is only a programme or software code. AI will only function within the parameters that it was originally set – whether they are right, wrong or their function rests upon the question of whether they are genuinely complete.

For a machine or AI programme to fully understand the human condition, it would either have to be living the complete human experience itself, with the abilities and understanding that we all have, or would to have been comprehensively and exhaustively programmed by a programmer or programmers together, who possess this inexhaustible knowledge themselves.

AI will not become sentient in the way that humans and other ‘living’ creatures can be.

Even the ultimate act of self-preservation suggested as possible by the AI movie legend that is Skynet from the Terminator films, was functioning based upon the limitations of its programming when it launched an all-out nuclear attack upon the world.

The real problem that we face with any and all forms of AI and Artificial Intelligence for the future, is the governance and the ethical boundaries or restrictions that any of its forms is able to operate under – even if that operation is itself carried out under the misused term ‘self-learning’.

No AI programmer or group of AI programmers have anywhere near the universal understanding of the human experience that would enable them to create a program that alludes to it, let alone covers the universality in difference of the way that humans actually think.

The reason for this is that no human exists on this planet today, who has the understanding or even the ability to understand everyone else’s experiences and the processes which make them think or have made them think, throughout the history of time.

As such, the inherent problem with any AI programming is that if the programmer, programmers or anyone who makes a decision upon what that AI can or cannot do, then do so in a subjective, biased or prejudicial way at any level at all – which will include their own lack of knowledge and understanding of what each and every other human being on the planet or that has lived and will live, faced with the same situation would do – they will not have provided the AI with every possible parameter that it could have.

AI is only as good as the information that is available to it. But that regrettably doesn’t mean that it wouldn’t or couldn’t kill a human being as a consequence of working with or within the limitations of what it could ‘learn’ or know.

AI accurately mirrors the limitations of the thinking of its creators. To be all knowing, it would not only have had to have been created by God, It would also have to be God for it to be so.

You can rest assured that if any human being alive possessed this ‘all knowing’, universal and arguably complete level of knowledge and understanding today, the last thing they would be planning to unleash upon the world would be any form of unethical AI.