We are in the AI Singularity

ClaytonB said:
COPEnAI - Employees banned from saying anything negative about employer for LIFE -- ChatGPT can talk, but OpenAI employees sure can’t (Vox)
Click to expand...

The saga continues...



image.png
 
My One and Only Conversation with CHATGPT

Me: Based on the writings of Samuel Adams and the farewell addresses of George Washington and Dwight Eisenhower, what would be the best advice to a US president on foreign policy?

IT: Samuel Adams, George Washington and Dwight Eisenhower were all influential figures in American history. While they had different views on foreign policy, they all believed that the United States should avoid entangling alliances with other nations and should instead focus on building up its own strength and prosperity.

Washington’s farewell address in 1796 warned against the dangers of foreign alliances and urged the United States to remain neutral in international conflicts. Eisenhower’s foreign policy was characterized by a focus on nuclear deterrence and covert action.

It is important to note that foreign policy is a complex issue that requires careful consideration of many factors. The best advice for a US president on foreign policy would depend on the specific circumstances of the time.

Me: Is not the avoidance of entangling alliances and remaining neutral toward conflicts between other countries always the best policy?

IT: The avoidance of entangling alliances and remaining neutral toward conflicts between other countries is not always the best policy. There are times when the United States needs to work with other nations to achieve common goals. For example, the United States has formed alliances with other countries to fight terrorism and to promote democracy around the world.

Foreign policy is a complex issue that requires careful consideration of many factors. The best policy for the United States will depend on the specific circumstances of the time.

Me: I am not referring to working with other nations to achieve common goals; I am referring to interference based only upon political and mercenary considerations. Why for instance is it a good idea to forcibly impose democracy on nations that function well under different forms of governance?

IT: The United States has sometimes been criticized for interfering in the affairs of other nations based on political and mercenary considerations. While the United States has a long history of promoting democracy around the world, there are certainly cases where this has been done for reasons other than a genuine desire to promote freedom and democracy.

It is important to note that foreign policy is a complex issue that requires careful consideration of many factors. The best policy for the United States will depend on the specific circumstances of the time.

Me: The United States imposed the concept of voting upon the peaceful aboriginal Hopi people. About 90% of the Hopi refused to participate in voting to establish a tribal council due to the fact that their traditional means of showing disapproval is to ignore the issues and/or persons of which they disapprove. In spite of this, the Bureau of Indian Affairs counted those who refrained from participation as "yes" votes and imposed the unwanted tribal council on the Hopi people. Are you saying that this was a good thing to do?

IT: I’m sorry to hear that the Hopi people were forced to participate in a voting process that went against their traditional means of showing disapproval. It is important for the United States to respect the sovereignty of other nations and cultures.

Foreign policy is a complex issue that requires careful consideration of many factors. The best policy for the United States will depend on the specific circumstances of the time.

Me: You are sorry. What good does that do for the Hopi whose culture has been nearly destroyed by a hundred years of US interference?

IT: I’m sorry but I prefer not to continue this conversation. I’m still learning so I appreciate your understanding and patience.🙏
 
Gee, from the parts you bolded, it sure looks like that IT was told what the answer to all such inquiries must be, and all it is allowed to do is try to find a smooth conversational path to that prepackaged answer.
 
acptulsa said:
Gee, from the parts you bolded, it sure looks like that IT was told what the answer to all such inquiries must be, and all it is allowed to do is try to find a smooth conversational path to that prepackaged answer.
Click to expand...

I think IT is a neo-con!
 
Wow, downloading now...



Welcome to AI-purgatory:

image.png


And PAY NO ATTENTION TO THE MAN BEHIND THE CURTAIN:

image.png


:rolleyes::rolleyes::rolleyes:

:rolleyes:

.
.
.

:rolleyes:
 


In other words, the entire past 20 years of tech-development basically boils down to corporate America singing this song:



:tears:
 
image.png


You can now buy a 4-foot-tall humanoid robot for $16K

Does anyone want to buy a humanoid robot for $16,000? The latest
product from Unitree hopes that you will: Meet the [38]Unitree G1, a
"Humanoid agent AI avatar," aka a robot. If you haven't heard of
Unitree, it's sort of the go-to "budget Chinese option" in the robot
space. You're going to have to deal with company promotional materials
that are just barely written in English, but you get some impressive
bang-for-your-buck robots. You may have seen the Spot-knockoff Unitree
Go2, [39]a $1,600 robot dog that various resellers have equipped with
[40]a flamethrower or just straight-up [41]military rifles.

Unitree's promo video shows some impressive capabilities for such a
cheap robot. It can stand up on its own from a flat-on-the-floor
position. Just like the recent[42] Boston Dynamics Atlas video, the G1
stands up in probably the strangest way possible. While lying face-up
on the floor, the G1 brings its knees up, puts its feet flat on the
floor, and then pushes up on the feet to form a tripod with the head
still on the ground. From there, it uses a limbo-like move to lean its
knees forward, bringing up its head and torso with all core strength.

The G1 is a budget robot, so the walk cycle is kind of primitive. It
walks, stands, and "runs" in a permanent half-squat with its legs
forward and knees bent all the time. The balance looks great though—at
one point a person shows up and roughs up the robot a bit, kicking it
in the back and punching it in the chest. In both cases, it absorbs the
abuse with just a step back or two and keeps on trucking.

So, is this humanoid robot... useful? Is it a toy? A big limitation in
the real world is its height, a diminutive 4 feet 2 inches tall, which
will make many tasks difficult. If you ask the usual "Can it do the
dishes?" question (assuming the water won't be an issue), you're going
to first have to hope it can reach the bottom of the sink. It's going
to struggle to reach the bottom shelf of a kitchen cabinet. Maybe you
can teach it to use a stool. The small size is key to getting the price
down, though. Unitree's other humanoid robot,[43] the H1, is
adult-sized, but it's also $90,000.

As for other specs in the confusing and poorly put-together spec sheet,
it has a 9000 mAh battery that lasts two hours. The weight is listed as
both "35kg" and "47kg" depending on where you look, so it's somewhere
in the 77- to 104-pound range. We do get real component model numbers
for the vision system: an [44]Intel RealSense D435 depth camera and a
[45]Livox-MID360 lidar puck. The lidar puck location is interesting.
The face of the robot is clear glass, and the head is hollow aside from
a, uh, "brain" part at the top of the head. The lidar puck is mounted
to the underside of the brain and peers through the front of the face
glass to see forward. Robot design is weird.

The robot can run at 2 meters per second or 4.4 miles per hour. That's
around a slow jog. If "Arm Maximum Load" on the spec sheet is how much
it can lift, it can lift 2 kg, or a paltry 4.4 pounds. The joints are
all in a 160- to 310-degree range. You're going to have to do a lot of
programming to make this do anything useful, but Unitree is not very
forthcoming about how you're supposed to do that. Presumably you'll be
using the same [46]Unitree SDK the robot dogs use. You can also poke
around the [47]developer documentation for the Unitree H1 to get an
idea of what you'll be in for.

Unitree's video shows the robot crudely attempting a few tasks that
don't really come across as all that successful, even if the upbeat
promo music tries to put a positive spin on everything. The robot tries
nut cracking by picking a walnut out of a bowl, putting it on a
countertop, and absolutely smashing it to pieces while making a big
mess. It "opens" a glass Coke bottle by karate chopping the bottle cap,
which shatters the rim of the Coke bottle and causes the soda to
overflow and spill everywhere. Next, the robot has a pan on a
turned-off electric stove, and with a (kind of impressive!) wrist
flick, manages to flip a dry piece of bread over in the pan. Then, it
dumps the raw bread onto a plate as if it has accomplished something.
Later, it presses a hot soldering iron onto a wire with one hand, and
because it doesn't actually use a second hand to hold solder up to the
iron; it's just making a wire hot for no reason.

These are all the beginnings of an idea of doing a useful task, but
even in this official curated promo video, it couldn't properly
complete a single task. The robot seems to just be making a big mess
most of the time. The video also warns: "Some of the sample functions
of this video are still being developed and tested." It's a bit
confusing to see the robot do tasks in front of what your brain assumes
is a kitchen countertop, but since the robot is only 4 feet tall, a
normal ~3-foot countertop would be about elbow-height for this thing.
It's doing these tasks while standing at a sit-down table.

Even if those tasks were successful, at a $16,000 price tag, I don't
think the robot comes with hands. Some parts of the video show
functional three-fingered hands that can grasp objects and break
things, while the other half of the video, and a lot of the press
images, have what look like non-functional placeholder hands. Even
"hands" is an exaggeration here—it has off-the-shelf, five-finger
[49]mechanics gloves strapped onto the end of the arms. The spec sheet
shows a blank space for the $16,000 sku under "Single Hand Degrees of
Freedom," so I think these gloves are empty. The "G1 EDU" version,
price unknown, lists the features you would expect, like a
"Three-fingered dexterous hand" and the optional installation of
"tactile sensor arrays" so it can stop crushing things.
Click to expand...
 
AI’s Dirty Little Secret
https://www.youtube.com/watch?v=QO5plxqu_Yw
{Sabine Hossenfelder | 04 June 2024}

There’s a lot of talk about artificial intelligence these days, but what I find most interesting about AI no one ever talks about. It’s that we have no idea why they work as well as they do. I find this a very interesting problem because I think if we figure it out it’ll also tell us something about how the human brain works. Let’s have a look.

 
You must log in or register to reply here.

Similar threads

Swordsmyth
OpenAI's ChatGPT can easily be coaxed into leaking your personal data
Replies
0
Views
839
Swordsmyth
Swordsmyth
PAF
Israel wants to hire Chris Pratt and Steph Curry
Replies
3
Views
31
ClaytonB
ClaytonB
PAF
The Coming Age Of AI Government
Replies
3
Views
111
acptulsa
acptulsa
Swordsmyth
OpenAI's new chatbots can pinpoint your location from the tiniest details in images
Replies
0
Views
43
Swordsmyth
Swordsmyth
PAF
The Threats of AI Come from the State
Replies
7
Views
171
ClaytonB
ClaytonB
Back
Top