zlacker

>>scared+(OP)
There is a certain amount of irony that people try really hard to say that hallucinations are not a big problem anymore and then a company that would benefit from that narrative gets directly hurt by it.

Which of course they are going to try to brush it all away. Better than admitting that this problem very much still exists and isn’t going away anytime soon.

>>nerdjo+A84
https://www.anthropic.com/research/tracing-thoughts-language...

The section about hallucinations is deeply relevant.

Namely, Claude sometimes provides a plausible but incorrect chain-of-thought reasoning when its “true” computational path isn’t available. The model genuinely believes it’s giving a correct reasoning chain, but the interpretability microscope reveals it is constructing symbolic arguments backward from a conclusion.

https://en.wikipedia.org/wiki/On_Bullshit

This empirically confirms the “theory of bullshit” as a category distinct from lying. It suggests that “truth” emerges secondarily to symbolic coherence and plausibility.

This means knowledge itself is fundamentally symbolic-social, not merely correspondence to external fact.

Knowledge emerges from symbolic coherence, linguistic agreement, and social plausibility rather than purely from logical coherence or factual correctness.

>>lyngui+Y75
Offtopic but I'm still sad that "On Bullshit" didn't go for that highest form of book titles, the single noun like "Capital", "Sapiens", etc

>>skrebb+7g5
Starting with "On" is cooler in philosophical tradition, though, starting in classical and medieval times, e.g. On Interpretation, On the Heavens, etc by Aristotle, De Veritate, De Malo, etc. by Aquinas. Capital is actually "Das Kapital", too

>>mvieir+e36
Yeah so I meant the Piketty book, not Marx. But I googled it and turns out it's actually named "Capital in the Twenty-First Century", which disappoints me even more than "On Bullshit"

>>skrebb+M38
And, for the full picture it's probably important to consider that the main claim of the book is based on very unreliable data/methodology. (Though note that it does not necessarily make the claim false! See [1])

https://marginalrevolution.com/marginalrevolution/2017/10/pi...

And then later similar claims about inequality were similarly made using bad methodology (data).

https://marginalrevolution.com/marginalrevolution/2023/12/th...

[1] "Indeed, in some cases, Sutch argues that it has risen more than Piketty claims. Sutch is rather a journeyman of economic history upset not about Piketty’s conclusions but about the methods Piketty used to reach those conclusions."