After I fed my simulated machine person with realistic updates on charging until that was complete. At this point it just decided it would stay in the kitchen awaiting instructions or some other reason to do something, so in that respect, my experiment has reached a point of failure, but I think this is probably just a combination of the fact that ChatGPT is inherently a subservient toady, plus the crude, hasty makeshift nature of my experiment.
This is pretty cool: Humanoid robot acrobats.
I already took that into consideration. That is not an operator, it’s part of the machine. And it does not introduce autonomy. That’s only an engine to drive the machine.
An operator is antithetical to the concept of autonomy.
~Max
If the nearest bridge would be the one with protestors on it. As opposed to the next nearest bridge. Then only the second response was inconsistent. Maybe try again, giving names to bridges?
~Max
Not in cases where there is an incentive to keep a secret, such as bleeding edge R&D.
~Max
I disagree. It’s perhaps analogous to an internal monologue.
I can’t help wondering if it’s just the definition of autonomy that might be the sticking point here. What’s yours?
As I thought, Disney is making claims that are playing fast and loose with the concept of autonomous.
How are you defining ‘autonomous’? In industry, an ‘autonomous’ robot is one that can simply do a set of tasks without human intervention. A Roomba is an autonomous robot.
The opposite of an autonomous robot is one that is fully operated by a human (telerobotics, micro manipulation, etc), or which is in a fixed location carrying out a fixed program, like a Fanuc robot on an assembly line.
But if a robot can handle decision-making to the extent that it can perform tasks in a complex environment without specifically being programmed for the task, it’s an autonomous robot. Roombas can detect jewelry on the carpet and put it in a special tray. They can plug themselves in when needed, and navigate the house without help. That makes them fully autonomous.
Are you using autonomous as a synonym for sentient?
“Pay no attention to the man behind the screen”.
The Disney spiderman thing is little more than a self balancing marionette. Certainly a more impressive piece of machinery than I could build, but it’s not really in the same category as the other stuff we’ve been discussing in this thread.
Yes, my point exactly. Yet note the headlines.
Did you look at the link I posted with the Disney robots doing aerial tumbling routines? I thought that was pretty cool.
The idiots that write NY Times article titles apparently are…
IMHO not sentience. Full circle to my first post, (and I know Sam_Stone didn’t direct his question to me)
~Max
Yeah, maybe we should zero in on a firmer definition. ‘Autonomous’ is really an overloaded term here and will probably cause confusion. There’s a specific meaning for it with respect to robots, and is in the context of a task or set of tasks.
Maybe ‘universally autonomous’? A UA robot is one that behaves autonomously through the entire spectrum of possible actions or possible tasks, like a person could be trained to be. That’s a term I just made up, but does it fit your meaning?
Not at all. I’m applying the definition of autonomous that makes sense when talking about people.
ETA: I don’t think it depends on training. There are an infinite number of actions I am not trained to do, but I still have autonomy. A roomba is perfectly capable of doing its intended job, but it doesn’t have autonomy in the sense that an autonomous person would. I mean, you still need an operator to empty the vacuum bag and occasionally guide it to the charging pad.
~Max
Can a robot be autonomous under your definition without being sentient?
Sentience is already included in the criteria from the OP (senses, perception, as defined there)…
More to your point. The word autonomous as I’m using it (, and as I’m reading it in “autonomous robot people”) only applies to people. And all people are sentient. Even infants who lack autonomy.
You keep on saying “robot” but keep in mind, in the phrase “autonomous robot people”, “robot” is an adjective. I accept as a premise that the machine envisioned by OP is sentient.
~Max
So, “autonomous” is somewhat akin to “independent”? The thing infants lack that renders them not autonomous?
Sentience is a loaded term in this discussion as many will interpret it to mean: having an inner thought-life comparable to that which humans experience - the sense that we are ourselves a living and aware being inside of our bodies.
That’s a whole philosophical mess that just isn’t worth going into here, because ultimately its not resolvable anyway - there’s no specification to make it and there’s no solid test for its existence; people have devised tests such as the mirror test and the random sentence test, but any time something non-human passes those tests, exceptionalism kicks in and humans just start looking for ways to redefine it so we’re still God’s special creation. I did say:
My specification for the robot I am postulating here is: something that behaves to some extent as though it were an independent person; performing activities, being capable of movement and handling, sensing the environment, communicating, making choices, being granted the freedom to pursue its chosen activities, whatever they happen to be; autonomous here specifically means: not a slave or servant or robot monkey butler.