The AI end user is using the tool as intended by the manufacturer: "Book me a yoga class". If the tool does something illegal/harmful it is the manufacturer's fault.
If the power hammer user uses the tool as intended but it explodes, it is the manufacturer's fault.
If the AI end user says "Hack the yoga website" or the power hammer user uses the tool to destroy something it's the end user's fault.
Pair programming requires a lot from you and your relationship with your pair programming partner.
When it's working I think it's absolutely great. Together we find the best solution in the fastest time with less errors along the way. I'm also less sloppy and more focused when someone constantly sees my work and vice versa.
At the end of the day you can celebrate what you've done, both feel responsible for the work and knows exactly how it works.
Works well if you're communicating while you're making changes in code.
At my work 20 people are changing the same wiki-page when we do heavy regression tests. Works like a charm (not everyone is writing at the same place in the document since we are testing different things and writing results at different sections).
If it's like the commercials in the movie the Truman show I think this could be a nightmare.
I think it all boils down to how (if?) this will be transparent to the user.
If the response suggests a sponsored product/service to the user it could either be the specific product/service that ChatGPT is biased to suggest or the whole idea of using a product/service.
As a user I would like to know both of these biases to understand that using any product/service for my need might not even be the best thing.
> Made it harder to be the "come on guys! we can win this" coach when we were down on points + bodies.
Would you say that paintball on that level is almost deterministic?
Compared to, for example, (American) football where momentum and perseverance can turn games around completely.
The AI end user is using the tool as intended by the manufacturer: "Book me a yoga class". If the tool does something illegal/harmful it is the manufacturer's fault.
If the power hammer user uses the tool as intended but it explodes, it is the manufacturer's fault.
If the AI end user says "Hack the yoga website" or the power hammer user uses the tool to destroy something it's the end user's fault.