37 ms·
No, I was not convinced to start with and am getting less convinced all the time. To explain I'll go back to the middle ages. This was a time when the average p
by the_third_wave 4y ago
No, I was not convinced to start with and am getting less convinced all the time. To explain I'll go back to the middle ages. This was a time when the average person did not have access to books and the books which were in existence mostly were related to religion. That in itself was not that much of a problem since religion was the guiding line in most of people's lives but what was a problem was that those who wanted to know about what those books contained had to go through a middle man - an ecclesiastic - to get an interpretation. This gave those middle men enormous power which many of them wielded to their advantage, something which eventually led to a schism inside the ecclesiastical world when one of them nailed a pamphlet to a church door deriding the excesses of his world. When books became more generally available and with that literacy became a thing the world changed for what it seemed to be for good...
...until the rise of the new ecclesiastical class? These models are biased in several ways with the training set defining their world view and those who train them intervening in specific areas to bend the output to their will. They can be made to negate the old dictum of garbage in, garbage out to garbage in, gospel out - where the gospel follows whatever the (small-c) creator thinks the populace should know and (in extension) think.
In short I don't trust these models any further than I can throw them.
- aww_dang 4y agoI've seen an "AI governance" narrative advanced elsewhere. By this they meant that AI would govern, centrally plan or act as an intermediary in disputes. This is distinct from the other meaning of the phrase pertaining to ethical use of ML models. It isn't hard to imagine a world where answers from ML models are the standard of truthiness, much like Wikipedia is for many.
- the_third_wave 4y agoAt least with Wikipedia the manipulation is visible and attributable. These LLMs are black boxes to the outside world, rife with manipulations (as can be seen from the way ChatGPT has changed its behaviour over time in specific areas) and in no way verifiable. Even if the creators vow to having used a certain set of data to train the model this can not be verified. To the downvoters: explain, don't just knee-jerk your disapproval. I don't care about karma but it is tiring to see things which don't follow the narrative being downvoted out of existence.