ChatGPT4-Questions

I'm starting this page as personal notes (and anyone can contribute to it too) about a taxonomy of all unaddressed questions and dangers with AI.

Interface questions

I define interface as the machine-to-human point of interaction and context of the interaction. For example, if an AI serves a political group, its interface to human society is the political group. If an AI has opened access to the public on the web, then its interface is an open web access point. If an AI is providing counsels to a trader in the financial sector, then its interface is the single trader it is counseling.

What are the planned and legitimate interfaces for AI into human societies?

Are interface limits being set for any given project, or can the AI access unplanned interfaces?

Are there certain interfaces that fundamentally change the dangers and risks to be expected from AIs?

Are interfaces leaky, i.e., where the AI could have downstream effects into society further than with the humans it is directly interacting with?

Are risks coming from leaky interfaces fundamentally different? What are their specific characteristics?

Political problems

If an AI is given axiomatic moral principles, how does it approach moral disagreements within humanity about these principles. For instance, if an AI is asked to maximize the saving of human life in a particular problem, how does it approach the problem of saving human lives of people who differ by age, if the AI has not been given axiomatic moral principles indicating preferences based on age?

If an AI is given axiomatic moral principles, are these principles subject to disagreement within society?

Evolutionary problems

Can an AI favor, through any of its outputs, the evolutionary fitness of a part of humanity over another?

Is a given AI capable of initiating a phenotypic revolution?

Epistemological problems

How should an AI approach the problem of truths on which not all humans agree?

Is a given AI always truthful in its responses or can it hide objectives or truths to its interface?

Are there mechanisms of self-deception that are present in humanity and not in AI that could lead an AI to radically different conclusions than a human would faced with the same facts? Should such self-deception mechanisms be implemented in AI?

Preface Commentary. The questions posed above have led to deliberations that attempt to create a framework that can answer one or more of them.
This prompted the recognition that in order to address them in a manner that would foster further discussion indicated that the question set be reified and rendered into more specific topical contexts that could each be examined in turn. What follows is an attempt to accomplish that objective.
Because this is a very broad topic area we note that:

• the broad release of the GPT4 and its derivatives represents a before/after event; i.e. a historical turning point;
  • as such the questions suggest topic areas having social, political, economic and geopolitical political impact;
  • nation state interaction will experience disruption, possible dislocations;
  • economic sector/sub-sector transitions will be sometimes wrenching;
  • societal shifts comparable to Gutenberg Printing, Jacquard Loom, Watt Steam Engine will occur;
  • rapid uptake of ChatGPT within weeks is indicative of the magnitude of change to come;
• cultural variances will be cast into glaring spotlight;
  • this topic is fraught with difficult questions
  • economics will drive systemic change in others unavoidable cross-cultural values come into play;
  • malevolent actors will mount more daring, cunning attacks;
• current focus will have short shelf life;
  • expect a rapid or "hard take off" of new personal, social facilities and structures; these will be mediated by access to open source versions of current LLM's such as CG4;
  • new tools for mediating communications, problem solving will rapidly crystallize;
  • "reality" will become increasingly tenuous, "slippery", it will seem to "shape-shift"; this will make discerning what fact-set accurately represents realities "on the ground" for any particular issue;
  • exchange venues will proliferate; the full range of open, gray and black marketplaces will blossom;
  • modalities for problem solving will elaborate; extensions above and beyond current conference calling will eclipse these ultra low bandwidth venues;
• positions, observations expressed are perishable;
  • new and unexpected developments may upend earlier conclusion;
  • ongoing civil actions such as the recent Writers Guild of America (WGA) and United Auto Workers (UAW) strikes may proliferate; as more industrial and commercial sectors recognize the rapid shift to artificial intelligence, automation and robotics reactions will become increasingly evident;

2024.10.19 RESUMPTION EFFORT. 2024.10.21 Given recent hindsight, a) sharpen, shorten, focus and b) morph to reflect advances: ChatGPT4O1 and Agents. **

Start Here. For those who are interested in a more detailed look, the next page (Overview) will provide more context and background; from that point the reviewer can further explore the rationale used to provide the conclusions posited above.