AI safety

  • 62 Replies
  • 6932 Views
*

keghn

  • Trusty Member
  • *********
  • Terminator
  • *
  • 855
Re: AI safety
« Reply #60 on: September 27, 2017, 08:54:30 pm »


 i have studied hard drug users. Human are tool using creatures. Using the there effects to get work done your be a god in whatever you believe in.
 like a teen age rap star.
 When a person uses it and relies on it for there main tool. They are super human for five to ten years and then after that they
are very very sub human for the rest of their LIFE. They need bigger and bigger doses just to be a normal person. And then that can be ever obtained.

 The subconscious mind will take over conscious mind if The A hunger is not dealt with. This the fragmenting of the mind. The sub conscious has much
greater greed and craving then the conscious mind, there are more drones and only one queen. And since primitive subconscious mind is closer to the bodies chemistry. If a person want to be a god of thinking or a rap star
then it will put chemical and hormones that make you become the opposite. Knowing you will double on the drug on till you get where you want to be.
 Once a conscious mined realizes this and tries to stop subconscious mind it has mean cocktail chemical in you blood to cause pain until it gets what it wants.
 Or if you get to near it, or other trigger, your arm and hand work on there own and you can not stop it. The junkie drones have kicked there queen and the hive is dis functional and normal hive can come in and take over.

Reward Hacking Reloaded: Concrete Problems in AI Safety Part 3.5: 



« Last Edit: September 27, 2017, 09:43:12 pm by keghn »

*

ivan.moony

  • Trusty Member
  • **********
  • Millennium Man
  • *
  • 1108
    • Some of my projects
Re: AI safety
« Reply #61 on: September 27, 2017, 09:59:22 pm »
I was just joking, I have to agree, drugs are off our limits, at least the most of them.
Dream big. The bigger the dream is, the more beautiful place the world becomes.

*

keghn

  • Trusty Member
  • *********
  • Terminator
  • *
  • 855
Re: AI safety
« Reply #62 on: September 28, 2017, 01:38:12 am »

What Can We Do About Reward Hacking?: Concrete Problems in AI Safety Part 4: 



 


MIT reshapes itself to shape the future
by Tyler (Robotics News)
Today at 12:01:15 pm
Tag... you're it...
by Art (AI News )
Today at 02:55:01 am
In need of a psychological coach?
by ruebot (AI News )
October 13, 2018, 01:33:37 pm
Leela Chess Zero
by Art (AI News )
October 08, 2018, 03:30:42 pm
Alpha GO Zero
by Art (AI News )
October 08, 2018, 02:48:32 pm
Detecting fake news at its source
by Tyler (Robotics News)
October 06, 2018, 12:00:55 pm
Phase Change Memory
by Art (AI News )
October 05, 2018, 01:21:59 pm
Model helps robots navigate more like humans do
by Tyler (Robotics News)
October 05, 2018, 12:01:40 pm

Users Online

59 Guests, 1 User
Users active in past 15 minutes:
Freddy
[Administrator]

Most Online Today: 99. Most Online Ever: 208 (August 27, 2008, 09:36:30 am)

Articles