What are the possible social consequences of training neural networks with artificially generated data?

Question

Machine learning models and, in particular, neural networks are trained with data often collected from the real world, such as images of real people.

Meanwhile, neural networks (such as GANs) are also used for data generation. Each year, they become better at this task to the point that even humans are not able to distinguish real-world data from the artificially generated one.

So it is possible that neural networks will start to learn with data that was generated by other neural networks, because it will look as real even for a human, but naturally will be not related to the real world.

1. Will it lead to some machine learning collapse?

2. Might it lead to some changes in human's perception of the world, because people get a very big part of their knowledge using computers, connected to the Internet?

3. Is anyone thinking about this potential problem?

Answer 1

We can already observe information bubbles on social media, where the circle is that the ML algorithms learn what content people like and give more similar content based on clicks and so on. From a single wrong click, you could enter a bubble and never come out if you don't take care or be aware.

This happens with humans, so the same may apply to computers. Checking backgrounds, like click saver for click titles as an ultimate example, would save the machines from circles too. Information is not different whether it is made in consecutive rounds by humans or machines.

For example, if you make a thesis, not necessarily all things you write are your own investigation and research. You use quotes, refer to papers, and so on. For information sources, you have to be careful if the source is respected and considered to be meaningful and correct.

The same carefulness needs to be applied to machine-generated content. To the responsible - the last responsibility is with the reader to believe or not what you're reading.

Answer 2

Will it lead to some machine learning collapse?

I wouldn't think so. Data is data. From the standpoint of automata, everything is ultimately reduced to a sting of bits. It may even be useful to be able to train AI's using CGI, for instance, in relation to automated vehicles. Not any different from humans using flight simulators.

Creating models and training AIs on them is useful, and a part of the contemporary AI landscape.

Might it lead to some changes in human's perception of the world, because people get a very big part of their knowledge using computers, connected to the Internet?

It already is. It's not only the false CGI content, but the scope of the search filter that dictates what information a websurfer gets. These results are controlled by algorithms, which evolve. Self-evolving algorithms may make the process more opaque. It definitely seems to be creating social problems already.

Is anyone thinking about this potential problem?

I'm sure there are papers out there on this subject. (Don't have time to search now, but I may do that and come back and amend with some articles and research papers.)

Two authors who are definitely thinking about this are Neal Stephenson and Hannu Rajaneimi. Stephenson addressed the "information unreliability" problem of the internet in Anathem. (It's not a major theme, but his ideas are quite insightful--Stephenson has a hard-science background, with a particular interest in computing.) Rajaniemi extends the ideas in the post-singularity Quantum Thief trilogy, where information and matter are interchangeable, and contains some very interesting ideas. (Rajaneimi holds two advanced mathematics degrees, which is useful in tackling a subject of such great complexity.)