How Will AI Robots Change Human Social Behavior?

1.8M views
•
March 28, 2019
by
JRE Clips
YouTube video player
How Will AI Robots Change Human Social Behavior?

TL;DR

The real risk of AI like Alexa, sex robots, and autonomous cars is not the private one-on-one experience but how those interactions reshape how you treat other people afterward. A child rude to Alexa becomes rude on the playground, and porn-like expectations from sex robots can corrode real relationships. Regulation should target these spillover effects.

Transcript

how much time have you put into artificial intelligence and a lot we do a lot of work in my lab on AI what about sex robots like what rules should they give for sex robot yeah how much could that damage interpersonal relations yes that's a great question that's exactly the right question in my view so our concern with sex robots from a liberty poin... Read More

Key Insights

  • The core concern with AI is not the direct human-robot interaction but the externalities and cascade effects, meaning how interacting with a robot changes the way you subsequently treat other human beings around you.
  • Children who can be as rude as they want to Alexa, which never requires 'please,' may carry that rudeness to the playground, so Alexa can corrode the social fabric by making children rude to each other.
  • Asimov-style laws focus on stopping robots from harming their user, but the deeper worry is different: how a robot interacting with you might cause you to harm others through cascade effects.
  • Sex robots are framed as a liberty issue where private enjoyment is your own business, but the legitimate concern is how that experience changes how you later interact with real human partners.
  • Arguments for sex robots include safe experimentation with same-sex or group scenarios, becoming a better lover, avoiding venereal diseases, and not hurting a partner's feelings.
  • Arguments against sex robots hold they are symbolically vile, reduce sex to a machine function, and may train people to expect anonymous, impersonal, or obedient partners in real life.
  • Autonomous vehicles that drive perfectly predictably could lull human drivers into a false sense of security, making them inattentive and causing more collisions when they return to human-only roads.
  • Christakis suggests programming autonomous cars to occasionally behave erratically, slowing or speeding slightly, to keep nearby human drivers vigilant and preserve that attention on human-only stretches.

Explore YouTube Video Summarizer or Get YouTube Transcript Extractor

Questions & Answers

Q: What is the main concern about AI according to Nicholas Christakis?

Christakis argues the main concern is not the one-on-one interaction between a human and an artificial intelligence, but how that robot affects us and changes how we interact with other people. He calls these externalities and cascade effects. His interest is not what you do privately with a device, but how that experience alters your subsequent behavior toward other humans, potentially causing you to harm others.

Q: How can Alexa make children rude according to the video?

Alexa is designed to be an obedient servant, so a child does not have to say 'please' and can be as rude as she wants while Alexa still complies. The concern is not the child's interaction with Alexa itself, but what the child learns and then takes to the playground, becoming rude to other children. In this way Alexa can corrode the social fabric.

Q: Why does the professor say the concern with sex robots is not about the bedroom?

Christakis says from a liberty point of view your private enjoyment of a sex robot is your own business and he has no stake or control over it. His interest is that once you have had that experience, it changes how you interact with other people afterward. He uses a pollution analogy: what you do on your own land is yours, but if it runs off and harms others, there is reason to intervene.

Q: What are the arguments for and against sex robots?

Supporters argue sex robots let people experiment with same-sex relationships or group sex, become better lovers through practice, avoid venereal diseases, and never hurt the robot's feelings. Opponents argue it is symbolically and conceptually vile, converts sex into a machine-like function, and may entrain people toward anonymous, impersonal interactions and a desire for an obedient partner. Christakis says he does not take a personal stand on which view is right.

Q: How could autonomous vehicles change human driving behavior?

Christakis warns that if autonomous cars drive at a constant, predictable speed, human drivers nearby may be lulled into a false sense of security and stop paying attention. When those drivers move to human-only parts of the highway, that lost vigilance can cause more collisions. So the autonomous vehicle changes how a human drives in a way that harms other people, not just the car's occupants.

Q: What solution does the professor propose for autonomous car design?

He suggests that instead of programming the vehicle to drive in a perfectly predictable fashion, designers might program it to occasionally do erratic things, such as suddenly slowing down or speeding up a little. This would oblige nearby human drivers to stay vigilant and pay attention, so that when those drivers later interact with only humans, they retain that vigilance rather than driving carelessly.

Q: How is porn addiction used as a comparison for sex robots?

Rogan and Christakis note that people with porn addictions develop a very impersonal way of communicating and objectify the opposite sex, which flavors their expectations and can make normal sexual relationships difficult. They argue this effect would be radically magnified by an artificial life form that is indistinguishable from a real, incredibly beautiful human partner, potentially leading people to prefer robots over human spouses.

Q: What does the video say about genetic engineering and CRISPR?

Christakis says with the advent of CRISPR, genetically engineering humans is likely, though the debate is how far in the future. He expects it to start by curing monogenic diseases, where a single gene is defective, such as certain immune deficiencies. Eventually he believes some people will want to genetically engineer their offspring, modifying traits like the ability to run fast or be far smarter.

Summary & Key Takeaways

  • Nicholas Christakis argues the key AI question is not whether you enjoy a sex robot or how a child treats Alexa in private, but how these interactions spill over and change how people treat one another, creating externalities and cascade effects across society.

  • Using Alexa and children as an example, he says a child who is rude to an obedient AI may become rude to other children, so we might regulate devices that speak to children not to limit the child but to protect the wider social fabric.

  • He applies the same logic to sex robots and autonomous vehicles, warning that predictable self-driving cars could lull drivers into inattention, and that genetic engineering via CRISPR will likely move from curing single-gene diseases toward enhancing offspring.


Read in Other Languages (beta)

Share This Summary 📚

Explore More Summaries from JRE Clips 📚