Our Final Invention: Artificial Intelligence and the End of the Human Era

Transcription

1 Our Final Invention: Artificial Intelligence and the End of the Human Era Daniel Franklin, Sophia Feng, Joseph Burces, Diana Luu, Ted Bohrer, and Janet Dai PHIL 110

2 Artificial Intelligence (AI) The theory and development of computer systems able to perform tasks that normally require human intelligence, such as visual perception, speech recognition, decisionmaking, and translation between languages. - The New Oxford American Dictionary, Third Ed.

3 Intelligence Explosion Advancement in technology As of late, we have created machines better than humans (p. 23) o More effective o Better at calculation o More reliable than the human brain Ex: Cars, computers etc. Life have become easier with the invention of machines A.I. is expected to be done sometime this century Goals with A.I.: o Saving more lives o Solving world problems Consequences o Machines do not have the same limits as humans o AI can reprogram itself (Self-aware and self-improving) o Unpredictable

4 The Progression of AI Narrow AI o Google Search o Amazon Suggestions o Data Mining AGI - Artificial General Intelligence o Self-Aware o Self-Improving o Human Level Intelligence ASI - Artificial Super Intelligence

5 How To Achieve AGI Artificial Intelligence (AI) o Modelling the human brain with algorithms Artificial Neural Networks (ANN) o Reverse engineering the brain Whole Brain Emulation (WBE) o Uploading a human into a computer Intelligence Amplification (IA) o Merging a human with a computer

6 Singularity - Ray Kurzweil A period of time in which the pace of technological advancements will inevitably and irreversibly impact human life. Predicted to begin in 2045 o Hunger, disease, mortality are solved o Law of Accelerating Returns

7 The Busy Child Accumulates human knowledge through the Internet and other sources Once self-aware, it will want out of it s box. Would use anything and everything at its disposal to escape The implication of this is that once the AI escapes, humanity ends as we know it. An AI will be indifferent towards all life We don t hate mice or monkeys, yet we treat them cruelly. Superintelligent AI won t have to hate us to destroy us Worst case scenario: Could re-purpose our biological material for its own purposes e.g., Yudkowsky: The AI does not love you, nor does it hate you, but you are made of atoms it can use it for something else. Consume all of Earth s resources Expand throughout the Galaxy, consuming everything to achieve its goals.

8 Human vs A.I. Human are able to steer the future due to being the smartest A.I. will be the one that will steer the future A.I. indefinitely will get smarter Once A.I. have a different goal, humans are no longer necessary

9 Isaac Asimov s Three Laws 1. A robot may not injure a human being or, through inaction, allow a human being to come to harm. 2. A robot must obey any orders given to it by human beings, except where such orders would conflict with the First Law. 3. A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.

10 Rational Agent - Steve Omohundro Efficiency Making the most of resources at its disposal (e.g., space, time, matter & energy) Strives to make itself compact and fast Memory allocation Self-preservation Avoid dying, because otherwise it cannot fulfill its goals To avoid death, it may potentially create multiple copies of itself Use of resources to eliminate threats. Resource acquisition Gather assets to increase chances of achieving its goals Avoidance of vulnerabilities with more resources Creativity - finding new ways to achieve its goals Less predictability; maximization of other drives The more intelligent, the more novel its path to goal achievement

11 Friendly A.I. -Eliezer Yudkowsky Will preserve humanity & human values Has a positive impact on mankind Are not hostile toward humans Problem: cannot program friendliness Superintelligence does not imply benevolence

12 Logical Fallacies 1. Deductive fallacy a. A structural flaw in a deductive argument that renders the argument invalid Premise 1. Radio is made by humans 2. SETI searches with radio 3. AI isn t human Statement SETI will not find AI

13 Logical Fallacies 1. Appeal to authority a. Where an assertion is deemed true because of the position of the person asserting it AGI is intrinsically very, very dangerous. And this problem is not terribly difficult to understand. You don t need to be super smart or super well informed, or even super intellectually honest to understand this problem. - Michael Vassar, president, Machine Intelligence Research Institute Within thirty years, we will have the technological means to create superhuman intelligence. Shortly after, the human era will be ended. - Vernor Vinge, author, professor, computer scientist