The 5 Safety Thoughts on AI

Today Google released an amazing Whitepaper titled Concrete Problems in AI SafetyIf you have the time to read it, I highly recommend it.

For all those who like to read the highlights, let me do that for you here :).

The Google Research Team published what they consider to be the five problems that will become very important as AI is applied in more general circumstances. The team called them the “forward thinking, long-term research questions” that are minor today but will become more important to address in future systems. (AKA Skynet or the Matrix or Abstergo).

Here they are:

  1. Avoiding Negative Side Effects: How can we ensure that giving the AI task will not disturb everything else around it while pursuing it’s goals. A Vacuuming AI that knocks over a vase as it goes faster to complete it’s goal faster.
  2. Avoiding Reward Hacking: How do we enable a cleaning robot to clean and not just cover up our messes?
  3. Scalable Oversight:  An AI system that gives feedback as it performs a task, it needs to use feedback efficiently to not keep bothering humans like the old Microsoft Paper Clip in Word.
  4. Safe Exploration: A mopping robot could be free to try different mopping techniques, but we wouldn’t want it to stick a wet mop into a live electrical socket.
  5. Robustness to Distribution Shift: A Robot that learns heuristics for a factory work floor may not be safe enough for an office.

Although these are great things to think about, there are more issues that need to be talked about and covered. I would suggest reading the White Paper if you want.

We believe in rigorous, open, cross-institution work on how to build machine learning systems that work as intended. We’re eager to continue our collaborations with other research groups to make positive progress on AI.

We maybe a little while off until we have the robots we see in most Sci-Fi films, but I believe Google Research is on the right path today!

Google on!

"Have you ever tried to count Hyperactive Children, while someone yells random numbers in your ear? Well that is what it is like in my head all day!" Trevor is apart of the robot revolution to slowly take over the human occupants of this Earth. By day Trevor is your above average Tech worker, tending his servers (Anton, Deckard, Anakin, Dimitri,and Scotty), and herding his 450+ employees. Yet at night...he sleeps, or as much sleep as he can get. Ideas always forming in his head with each new marvel and innovation to arrive. Each thing helping him to decide to help his robot overlords more and more. Trevor holds a few different designations: -Google Certified Administrator -Microsoft Technology Associate -Microsoft Specialist: Configuring Windows Devices -Microsoft Certified Professional -Microsoft Office Specialist He loves reading, working on SMART homes, and writing about his passion for Google products and innovations. He and his wife live with their 2 dogs and 3 cats in their quiet little beach town. Feel free to reach out to him!

Leave a Reply

Your email address will not be published. Required fields are marked *