Yes, John, you are right that people are developing and building them. It is also true that AI is now building AI, just as the increasing power and sophistication of computing has for some time allowed computers to themselves generate code to make things happen. In many regards, I suggest that AI is simply a continuation of that.
What can we do? I'm no expert and have no 'clever' answer to your question, not least because it is a general one and could be asking about any or all aspects of AI and its development, use, control, benefits, problems or issues and so on.
The best I can offer you right now is some general categories in which action, in my opinion, ought to be taken if AI is to serve people rather than people serving AI.
DISCUSS AS WIDELY AS POSSIBLE NOW and include not only obvious stakeholders but the poor, the ignorant, the unworldly, the young and the aged, and in fact everyone who seeks to or ought to understand the implications of AI, whether beneficial or detrimental to human well-being.
Some possible considerations:
* Create moral and ethical PRINCIPLES for developing AI, preferably, to which most if not all can agree
* Detemine what AIMS are appropriate for AI and what are not. (Allows allowing for flexibility and adaptation in response to that of which no has yet thought.
* Recognise that AI must serve PEOPLE FIRST rather than people serve it.
* Devise a means of MONITORING all AI development and implementation.
* Ensure that decision makers such as politicians, CEO's, Academics, and other such influencers are fully informed so that LEGISLATION can keep pace with innovation in AI.
* PROTECT CONTENT that is collected to provide the data used by AI to feed its algorithms (if that's the right term).
* Require MANDATORY LABELING of AI generated or manipulated product so that users of its results are aware that what they read, see or use is only indirectly under the control of actual human beings.
I accept that these are my own, less than fully informed or knowledgeable feelings and thoughts of potential ways in which possible dangers of AI might be minimised.
I also accept that it may be difficult or virtually impossible to ensure 100% agreement, implementation or adherence to some of them, whether on a local, national or worldwide level.
As you are clearly aware, AI is here. It will not go away. There are those who will use it to improve life and benefit many and there are those who will use it to enhance only the life of themselves or their favoured group. There are still others who will use it for intentionally nefarious means and those will unintentionall use it, not realising that what they are doing may have catastrophic results for some or all.
It may, indeed it will sound alarmist to many but my firm belief is that AI is potentially as potent a threat to humanity and its development and use as complex an issue for resolution as was and is atomic energy.
The horse has bolted on nuclear energy and it may have done so on AI but perhaps on AI we still can ensure that the barn door stays closed unless and until we can let the horse out in a safe and controlled way.
...
I'm sorry, I have no connections in this field that could be useful to you, indeed I have virtually no connections at all for I don't and haven't networked. Achieving and maintaining competence and currency in my work has always been my primary concern. However, I wish you well for the future and hope that you achieve what you'd wish and use your expertise wisely and in the interests of a more equitable society.
Take care. Stay safe. ☮️
