Non-political things that make you say...WTF?!?! (Welcome To Florida!) **NOT CV RELATED** NO TALKY!

You’re too limited in you’re definitions of “robot”.

 

Whether a mechanical device controlled by computers or a back end systems application sorting through submitted resumes, the decisions are made based on parameters created by humans.

Any claims to the contrary are someone trying to skirt responsibility for the results of those decisions.

 

This reads suspiciously like a robot wrote it.......

 

Ask Alan Turing....

 

https://youtu.be/skGbcOSj5D0

 

You say this like it's a bad thing. I'm not even going to argue if this is "robotics", the reality is that the "robots" are doing the exact same job, using the exact same metrics that were used before they were, they're just better at their job than their human counter parts. They can process applications in seconds, not hours. They can approve or deny someone based upon their actual worthiness instead of say race, gender, religion, or their hairstyle and they make a lot fewer mistakes while doing all that.

 

One of the best deciders of “worthiness” is home zip code. It totally reflects “income”, which in its turn reflects “race”, “gender”, etc. As long as you live in a upscale zip code all is peachy. As far as automation making fewer mistakes, automation repeat mistakes faster than any human or team of humans.

 

And as someone said earlier, that's not automation's fault. Automation does what it's programmed to do. Its not making mistakes, it's doing what the programmer programmed it to do.

 

So you play the “its not my fault I was doing what I was told” defense. Automation/technology is responsible for a large displacement (of people) and enables a even larger displacement in the future.

I did get you to back off the its “just better” statement.

 

I think the point is that it is humans that designed the AI models and trained them with data. PEOPLE are responsible, not automation and technology.

There is lots of pending legislation across several states to help thwart this. In the AI/ML field, eliminating bias and being able to prove that a model is not biased is a major focus right now. One of the main issues is that a model that is shown to not have bias at a certain point in time may learn to be biased over time due to the data fed in to it. Being able to monitor and curate this data is even another potential area where bias can be introduced.

Best to stay away from things like zip code to train models that make decisions that are impactful to individuals.

 

No, I play the robots currently used are incapable of thought defense. The majority of them operate off Gcode. I'll simplify Gcode in case you don't know, it is nothing more than coordinate instructions. Rotate servo1 5 degrees, Rotate servo2 15 degrees, activate solenoid3, rotate servo 2 45 degrees, deactivate solenoid3, and then the gcode loops 950,000 times.

As for automation displacing people. I find this funny when I hear this, usually from people with no direct experience with automation and where it is being utilized. I've worked in the manufacturing industry for nearly 2 decades. And in that time I have NEVER seen a shop that was fully staffed. Quite to the contrary, almost all shops struggle to find people, let alone qualified/capable people, despite good wages and good benefits being pretty much standard throughout the industry. The shop I am currently at uses 3 robots, and has 2 more ordered. The truth though is that they would prefer to hire people. Robots often take weeks or often times months to setup for every job. Rotating servo1 5 degrees as in the example above might only for 90% of the time. The other 10% the machine is crashing and someone that was supposed to be replaced by that robot is going to have to go into the gcode and change that 5 to a 5.05. It would be much faster to teach a person to do the same job because they are capable of thought and reasoning, but you can't train what you don't have and can't find. And you can't displace someone that doesn't exist.

The inevitable path this discussion leads down is AI. Current AI is really nothing more than pattern recognition. Lets take google assistant for example. Google is trying to make it so that it can call a restaurant and order a pizza for you. Most people think google assistant is "thinking", but the reality is if you take 1,000,000 calls to pizzahut, 95% of those calls are going to be nearly identical. So the assistant calls pizzahut, someone answers and says "thank you for calling pizzahut, how may I help you", the assistant then converts that speech to text, searches through it's 1,000,000 calls and finds that 90% of the time the response is "i'd like to order a pizza". It's really nothing more than a probability calculator with the ability to look through a large database in seconds and recognize patterns. That however is leaps and bounds away from critical, creative, independent thought. We are decades if not centuries away from AI being able to develop things like new technology from scratch, let alone having an idea it's self.

 

Speaking of Robots

On the home screen here at TW there is a small window to the right entitled "Members Online Now".
It appears to keep track of how many people are on the site; how many are members and how many are guests.

I just noticed tonight that there is also a category for how many Robots are Online as well ! ?
At this very moment it says there are 434 of them !!

Can you guys tell me what that is all about ?

 

Webcrawlers.

 

Is says on the can it's 25ml more than 250ml. It just so happens to be in the same size can as 300ml. Same size bags of chips don't have equal measurements but could charge more or less for either. Marketing baby! I don't know why I bothered to write any of this. Does having this information even matter?

 

https://youtu.be/G83kJF_HaIk?t=223

 

Dude hits VTEC VVT-i at 00:16, unleashing all 204 hp

https://youtu.be/H7MmmaGgxvM

In other news, the Hilux gets paddle shifters!

 

Fucking imbecile.