If you believe this study [1], humans can guess party affiliation at least slightly better than random chance from images alone.
Or [2] is an (unscientific) exploration from the other direction, prompting image generation models to make images of republican and democrat voters, with very different results
Presumably, everything you have done publicly (and hence your personality) exists somewhere in the big Google neural network. It gets compressed into one of the many billion weights. It might be hard to decompress it into useful information. But it is there nonetheless. Just showing your face might trigger and activate some layers in there.
That's sometimes possible (e.g. the "Trump woman" look, or certain "I know it when I see it" stylistic cues mainly displayed by progressive women that I can't really articulate). Polarization has turned political alignment into subculture, and members of subcultures often dress certain ways (and not necessarily consciously).
I'm referencing it as being possible, however I didn't share benchmarks because candidly the performance would be so slow it would only be useful for very specific tasks over long time horizons. The more practical use cases are less flashy but capable of achieving multiple tokens/sec (ie smaller MoE models where not all experts need to be loaded in memory simultaneously)
what happens if I test this tool by installing some packages and then remove (the tool)? will I still be able to use Homebrew to manage these new packages?
Basically it's about the quality of the code output and scenarios flexibility. Recorders that don't have any AI layer on top produce:
- Brittle selectors
- Dump entire scenario into a single file
- Do not allow any logic besides replaying your steps
Qure writes clean code with properly reusable abstractions like Page Object (or whatever do you have in your existing automation repository). For the team it means less flakiness and less time spent on maintenance
And if we consider AI-powered tools, they require much more steering from the human perspective: describing test in text, forcing agent to follow specific guidelines (e.g. for selectors), asking agent to refactor stuff after main scenario is working. Qure agent handles that out of the box itself
