Will robots ever be able to see colors the way we do?

Don’t digital cameras “see” color on the sensor, and identify it with their processor?
My point is, it’s sort of already happened.

Different animals see colors in different ways. I think it is dogs that are essentially colorblind; others see a broader spectrum than most humans. And humans are not uniform in how they see color.

But just as I cannot tell exactly if you “see” the red that I do, optical readers can already see colors just as well as you.

Yes, but could they write a poem about the experience?

The nature of perception has been a philosophical question of long standing that is not likely to be answered any time soon. Here is a link It is kind of long, but I think it is of interest to skim it to get an idea of the arguments involved.

You’re asking about two different things – two very different things.

On the one hand, calibrated instruments built to do it can perceive differences in wavelength far more exactly than humans ever could. So that question has been answered.

But then you followed that up with whether machines / instruments / computers / robots can develop the same emotional response to color that humans have. (Which also sort of presupposes that one human’s emotional response to a particular wavelength will mirror another’s.)

Those clearly aren’t the same question.

Yes, we already have sensors that can discriminate and determine color to a great degree. It is only a matter of applying this to robotics, which I believe we already have in the latest mini and nano drones that are being tested now by the military to positively identify targets and communicate kill recommendations to a remote missile system. They can be dropped by fighter/bomber aircraft from safe heights, swarm in the hundreds, communicate with each other thousands of times a second, follow instructions in real time from a remote central command a continent away and are so fast that regular cameras cannot pick them up, nor can radar due to their size. They can also be weaponized.

Dogs are not colorblind in the sense that they see more than just the gray scale. However, the color range they perceive is limited compared to the spectrum we see. To put it in very basic terms, the canine color field consists mostly of yellows, blues, and violets.

We determine this by the distribution, density and ratio of color photoreceptor cells—the rods and cones found on the inside surface of the mammalian eyeball. Cones function best in relatively bright light, as opposed to rod cells, which work better in dim light. Dogs have less rods and cones than we do, but they do have them. Cats have many more rods than cones, which explains their excellent night vision.

A tech version of rods and cones have been essentially replicated.

Will humans ever go to the moon?

I am not asking two different questions. We do not perceive color as wavelengths or frequencies. We do not see either red or violet, on opposite ends of the visible spectrum, as being more of anything than the other. Similarly, we do not sense smell on a linear scale. There is no single scale for comparing chocolate, vanilla and strawberry.

So far science does not have a clue as to the inherent nature of our perceptions. If we are clueless as to such a fundamental aspect of our being which we share with other creatures, think of how much more difficult it would be to create a robot that has a sense of self, which we share with at most a handful of other species. A simple test of this would be to give a robot a mirror and, without prior specific programming, have the robot recognize its reflection. Humans and just a few other species can recognize themselves in a mirror. Robots still can’t pass the Turing test.

I think color perception has more to do with how we process information than how we receive it. I’m sure we could, in theory, hook a computer up to an eyeball and print out color values as frequency rates. After all, the three types cones in the back of the eye are simply receptors that respond to certain frequencies of light (red, green, and blue for humans.) What I mean is, I’m not sure there’s anything inherently special in the way our brains receive color information… other than that’s how it happened for us. In fact, octopuses don’t have cones at all, yet see colors quite well.

What I mean is, even if a robot receives color information that’s nothing more than number values from low to high, I think it could still perceive that color. It would just depend on how its “brain” processes the information. It doesn’t have to process the numbers as values of greater or lesser quantity. The numbers would just be a way to organize the information, like an alphabet organizes letters.

However, I do think a perception of color like what we’re familiar with would require some sort of consciousness. In other words, I think the larger question surrounding this one is, “Can computers become sentient?” As far as I know, even the most sophisticated AI is still, at its core, rather “artificial”… that is, it’s still as much blind machinery as any of our other computing. We still have a ways to go.

Brains process the information they receive from sensory inputs by constructing an image of reality (source: my brain). Other animals have more or less (or different) input from various senses, but I think it’s more than reasonable to assume they perceive the world in ways that would be familiar to us—we have the same kinds of senses and same kinds of neural structures, after all, even if they vary from species to species.

What color would look like to a sentient robot, I can’t fathom. I guess most of it depends on what color information the robot receives (what spectrum), and how it’s designed to cognize of that information. Hopefully we also give it ways to communicate its inner experiences with us, so we can get some idea… (so that it could write that poetry, I mean.)

I’m not even sure you see colours the way I do so…

They will look up at a starless night sky, and will be confused why you humans can not see this beauty

Robots could see colours the way we do if they were like us but they’re not. Red is meaningless to them except as a wavelength of light. The appreciation of colour is in the mind of the beholder and robots do not have minds. Maybe someday they will but not today and not soon.

Exactly the point I was trying to make. Computers can do some extraordinary things and maybe some day we will reach the singularity, but I don’t think that time is near, though it is not too early to contemplate the possible consequences..