Great work here, but I do feel that the only important observations in practice are those about reasoning. To the extent that obtaining visual information is the problem, I think the design of language models currently is just not representative of how this task would be implemented in real robotics applications for at least two reasons:

1. The model is not using anywhere near all of the information about an image that it could be, as language models which accept image data are just accepting an embedding that is far smaller (in an information theoretic sense) than either the original image (probably) or the activation spaces of a large independent vision model (definitely).
2. Even if we accept the constraint of using less data than we have by converting to an embedding and dropping it in line with other text tokens as these models mostly do, a real system would have the opportunity to consume many images if it was attached to physical machinery that was used to capture them and could eventually get far more data than the embedding provides.
   
   ——
   
   I believe a hypothetically “real” system of this kind would almost always be implemented as a powerful vision model scaffolded on to a language model that directed it as a tool to perform an analysis progressively, operating on this data in a manner that is very different from the one-shot approach here. While these experiments are very helpful, I personally am not updating much toward it being very hard to pull off - except where the evidence is about the reasoning after the fact of understanding the image data correctly (which obviously will always be important even if vision is provided as an iterative tool of much higher bandwidth).

This is also how I interpreted it.

It would be great if there were more options. I would absolutely leave my current job, and bring my ML experience with me, to a role in AI safety. I would be okay to take a pay cut to do it. This doesn’t seem like an option to me though, after a brief bit of searching on and off over the last year.

Great essay. Though I would note that “price gouging” usually refers to the scenario described: when the change in price is possible only by virtue of seller market power. I think the term is misused enough that it makes sense to present the example as is, but I would call it a pretty basic error in terminology for the scenario, absent seller market power, being referred to as price gouging.