> More opportunity on the table.

Hate to disappoint you, but as the models get better, and eventually deliver the results, you won't have to wait a microsecond until the masses roll in to take advantage.

An attack on the result might be interpreted as an attack on the cause. Maybe a system's purpose is what is does, after all.

The Academy Awards? Who the heck cares for those?

The only thing they say is "the powers that be in Hollywood chose to highlight your product", but surely the information content regarding quality is as close to zero as it is possible to get.

It's nice to see some good news now and then.

Dunno, I'm happy to pay for API access based on token usage. I'd never so much as look at flat pricing, but maybe that's just me.

Exactly because the task is harder if the variable name does not contain any information is what makes training like that a good idea. It forces the LLM to pay attention to the actual code to get it right, which in training is a Good Thing (TM).

Let me add the part where politicians do not give a fig about their rhetoric from before the election as soon as they get into power.

Politicians and their buddies scared of having lost control of mass media.

It's easy to find the company (https://www.xreal.com/us/air/) but they sell different models, and I'd be interested in gp's model and user experience.

Male same-sex relationships have the lowest numbers of abuse between all four categories, btw.