Center for Analysis and Design of Intelligent Agents

This page is read only. You can view the source, but not change it. Ask your administrator if you think this is wrong.
[[/public:t-709-aies:AIES-24:main|DCS-T-709-AIES-2024 Main]] \\
[[/public:t-709-aies:AIES-24:lecture_notes|Link to Lecture Notes]]

\\
\\

====== FUTURES STUDIES ======
//Knowing the future//

\\
\\

==== Futurology ====

|  What it is  | Futurology is what we call "studying the future".  \\ Systematic prediction of what may happen in the future, and when.      |
|  Why it matters  | Allows preparation for what may come.      |
|  How it Works  | A bit like quantum physics: You can either say **what** will happen, or **when** something might happen, but not both at the same time.      |
|  The Next 20 Years  | To understand how AI may impact ethical concerns, and what the ethics of AI will become, we must look to the future. \\ That, however, calls for a crystal ball.   |
|  Further Readings  | [[https://en.wikipedia.org/wiki/Futures_studies#:~:text=%22Futurology%22%20is%20defined%20as%20the,a%20new%20science%20of%20probability.|Futures Studies]] on Wikipedia    |


\\
\\

==== Crystal Ball ====

|  Our Crystal Ball  | is composed of parts that allow studying certain key factors in isolation.   || 
| | We can arm ourselves with: ||
| |  1.  | Remaining open questions in AI. |
| |  2.  | Understanding the drivers of "progress".   |
| |  3.  | Identifying and mapping out key causal factors of change.     |
|  And  | use them to dissect, study, and analyze their nature, behavior, and interaction.    ||

\\
\\

==== Three Claims About the Future ====

|  #1:  | Rate of progress in automation through AI depends on rate of progress of basic research in AI ||
|  #2:  | Rate of progress in AI basic research depends on when these four mental processes will be automated:   ||
| |  a.  | **General autonomous learning**     |
| |  b.  | **Empirical reasoning** over autonomously learned knowledge     |
| |  c.  | Creative autonomous **goal generation and management**     |
| |  d.  | All three **unified in a single system**    |
|  #3:  | Automation with AI will be at least as impactful as automation in the last industrial revolution. ||

\\
\\


==== Automating Intelligence ====

|  What AI is Really About  | Automating intelligence.    |
|  Prerequisites  | Fundamental knowledge about how to automate intelligence is a pre-requisite for applying automated intelligence government, business, health industry, and all other areas of life.    |
|  Home of Basic Research  | Basic research is primarily performed in academia.  \\ Hence: progress in the application of AI depends on progress of basic research in academia.   |
|  "Rate of Progress"  |  The speed of basic research depends on human ingenuity, funding, enabling factors.   |

\\
\\

==== "Rate of Progress" ====

|  What it Means  | Essentially means the speed of “creating new knowledge”.  |
|  Creating New Knowledge  | Main producer of new knowledge is **systematized research** - applied and basic. \\ Scientific/technological research uses a system based on addressing forward-looking questions, from simple, e.g. how to manufacture something faster, to extremely difficult - curing cancer, implementing intelligent machines, understanding the cosmos.   |
|  Basic research  | Primary focus on fundamental questions. \\ Basic research in AI: Automating more (human) thought processes.  |
|  Applied research  | Primarily profit-, control-driven.  \\ Applied research in AI: Applying thinking machines to practical tasks.  |
|  Key Dependencies  | Applied research relies on progress in basic research. \\  Thus: Basic research drives applied research.   |

\\
\\

{{/public:t-709-aies:future-march-of-ai.png}}


\\
\\

==== The Fourth Industrial Revolution is the AI revolution ====

|  4th Industrial Revolution  | Has just begun.   |
|  Economics  | Under current market mechanisms, whatever is profitable to automate with AI will be automated with AI. \\ Power of AI will improve significantly over the coming years and decades.   |
|  What to Do  | Governments, special interest groups, and society at large must ensure that we benefit from AI, and don’t succumb to its negative side-effects.   |
|  Quite possibly jeopardized by AI  | Democracy, the law, nation states, our monetary system, peace, our way of life.   |

\\
\\
\\
\\
//2024(c)K.R.Thórisson//