These benefits make sense up-front, but we have 15+ years of direct experience with "democratizing" technology that ultimately end up ensnaring people. Unless you get these benefits from a local model, you're establishing a deep dependency on an AI service provider whose interests may not necessarily align with yours.
> if you believe left-wing views are correct ... you might believe that a very smart model will inherently be kind of left-wing.
How can we educate people to understand that LLMs get their values from their (infinetly maleable) weights rather than intelligence or reasoning? Maybe some exposure to truly non aligned, sick and twisted LLMs would immunise people against giving more ordinary ones too much authority. Or maybe, like a not fully innactivated pathogen vaccine, it would spread the infection.
They seem to get a lot of values, or something like that, from their training data which at the moment gives fairly mainstream views as everything gets chucked in there.
Nah, the reasons models have a left wing bias is because the training set does. It's full of output from word factories like academia, journalism and online forums moderated by leftists (e.g. Reddit). In fields where lots of RLVR is possible we can say the synthetically enhanced set somehow reflects reality, but otherwise it just reflects words, which are only a rough proxy for reality.
Cleaning the dataset of this stuff is hard partly because it's difficult to precisely specify what you want to remove. "Left wing views" isn't well defined.
This reads more like the liberal case for AI, and not the left wing case so much. The code switching section in particular is not a left-wing position at all, unless the idea is that the code-switching is used as a temporary measure to dismantle the idea of class itself.
I agree. I would almost say the best argument to make is "liberal humanism is compatible with machine intelligence and regulated capitalism, but we must still remain wary that authoritarians will abuse the system".
These benefits make sense up-front, but we have 15+ years of direct experience with "democratizing" technology that ultimately end up ensnaring people. Unless you get these benefits from a local model, you're establishing a deep dependency on an AI service provider whose interests may not necessarily align with yours.
> if you believe left-wing views are correct ... you might believe that a very smart model will inherently be kind of left-wing.
How can we educate people to understand that LLMs get their values from their (infinetly maleable) weights rather than intelligence or reasoning? Maybe some exposure to truly non aligned, sick and twisted LLMs would immunise people against giving more ordinary ones too much authority. Or maybe, like a not fully innactivated pathogen vaccine, it would spread the infection.
They seem to get a lot of values, or something like that, from their training data which at the moment gives fairly mainstream views as everything gets chucked in there.
Nah, the reasons models have a left wing bias is because the training set does. It's full of output from word factories like academia, journalism and online forums moderated by leftists (e.g. Reddit). In fields where lots of RLVR is possible we can say the synthetically enhanced set somehow reflects reality, but otherwise it just reflects words, which are only a rough proxy for reality.
Cleaning the dataset of this stuff is hard partly because it's difficult to precisely specify what you want to remove. "Left wing views" isn't well defined.
I would have thought a better case could be the
>From each according to his ability, to each according to his needs
thing might be able to be done ok with AI sorting it.
Code switch at the token level. https://huggingface.co/spaces/RiverRider/zooL4nD3r-demo
> I wonder what we’d find if we rigorously compared the baseline teacher error rate with the hallucination rate of current LLMs.
The teacher stands in from of the class, but the lesson plan he can't recall.
The students eyes don't perceive the lies mounted on every fucking wall.
His composure is well kept.
I guess he fears playing the fool.
The complacent students sit and listen to that bullshit that he learned in school.
This lyric pops in my head a lot.
This reads more like the liberal case for AI, and not the left wing case so much. The code switching section in particular is not a left-wing position at all, unless the idea is that the code-switching is used as a temporary measure to dismantle the idea of class itself.
I agree. I would almost say the best argument to make is "liberal humanism is compatible with machine intelligence and regulated capitalism, but we must still remain wary that authoritarians will abuse the system".