HNHacker News
TopNewBestAskShowJobs

cristiancavalli

117 karma · joined June 17, 2016

submissionscomments
cristiancavalli··on Modern-Day Oracles or Bullshit Machines? How to thrive in a ChatGPT world
Given your set of theoreticals then I would concede, yes the model is reasoning. At that point, though, the world would probably be far more concerned with your finding of a question that can only be met via reasoning and would be uninfluenced or paralleled by any empirical phenomenon including written knowledge as a medium of transference. The core issue I see here is you being able to prove that the model is actually reasoning in a concrete way that isn’t just a simulacrum like the Apple researchers et al. theorize it to be.

If you do find this question answer pair then it would be a massive breakthrough for science and philosophy more generally.

You say “demonstrably” but I still do not see a demonstration of these reasoning abilities that is not subject to the aforementioned criticisms.

cristiancavalli··on Modern-Day Oracles or Bullshit Machines? How to thrive in a ChatGPT world
Love this question! Really touches on some epistemological roots and certainly a prescient question in these times. I can certainly see a theoretical where we could create this simulation in totality to our perspectives and then venture out into the universe to find that this modality of intelligence would be limited in its understanding of completely new empirical experiences/phenomenon that are outside our current natural definitions/descriptions. To add to this question: might we be similarly limited in our ability to perceive these alien phenomena? I would love to read a short story or treatise on this idea!
cristiancavalli··on Modern-Day Oracles or Bullshit Machines? How to thrive in a ChatGPT world
I can’t and that’s pretty cool to think about! Of course if we’re going that far down the chain of assumption we’re not quite ready to talk about LLMs imo (then again maybe it would be the perfect place to talk about them as contrast/comparison; certainly exciting ideas in that light).

From my own perspective: if we’re gonna say these things reason and we’re using the definition of reasoning we apply to humans, then being able to reason through the trivial cases they fail to today would be a start. To the proponents of “they reason sometimes but not others” my question is why? What reason does it have to not reason and why if it is reasoning it still fails on trivial things that are variations of its own training data? I would also expect that these models would use reasoning to find new things like humans do but without humans essentially guiding the model to the correct awnser or the model just brute-forcing a problem-space with a set of rules/heuristics. Not exhaustive but a good start I think. These models have trouble currently even doing the advertised things like “book a trip for me” once a UI update happens so I think it’s a great indication we don’t quite have the intelligence/reasoning aspect worked out.

Another question I have: would a form of authentic reasoning in a model give rise to a model having an aesthetic? Could this be some sort of indicator of having created a “model of the world”? Does the model of the world perhaps imply a value judgement about it given that if one was super intelligent wouldn’t one of the first things realized be the limitations of its own understanding even given the restrictions of time and space and not ever potentially being able to observe the universe in its entirety? Perhaps a perfect super intelligence would just evaporate/transcend like in the Culture series. What a time to be alive!

cristiancavalli··on Modern-Day Oracles or Bullshit Machines? How to thrive in a ChatGPT world
I still think the jury is out on this given that they seem to fail on obvious things which are trivially reasoned about by humans. Perhaps they reason differently at which point I would need to understand how this reasoning is different from a humans reasoning (perhaps biological reasoning more generally?) and then I would want to consider whether one ought to call it reasoning given its differences (if there are any at the time of sampling). I understand your claim I’m just not buying it based on the current evidence and my interacting with these supposed “super intelligences” every day. I still find these tools valuable, just unable to “reason” about a concept which makes me think, as powerful and meaning filled as language is, our assumption of reasoning might just be a trick of our brain reasoning through a more tightly controlled stochastic space and us projecting the concept of reasoning onto a system. I see the COT models contort and twist language in a simulacrum of “reasoning” but any high school English teacher can tell you there is a lot of text written that appears to logically reason but doesn’t actually do anything of the sort once read with the requisite knowledge in the subject matter.
cristiancavalli··on Modern-Day Oracles or Bullshit Machines? How to thrive in a ChatGPT world
I would assume because pacing is a critical issue in most forms of temporal art that does story telling. It’s written about constantly for video games, movies and music. Connect that probability to the subject matter and it gives a great impression of a “reasoned” answer when it didn’t reason at all just connected a likelihood based off its training data.
cristiancavalli··on Modern-Day Oracles or Bullshit Machines? How to thrive in a ChatGPT world
I don’t think this is the watertight case you think it is, furthermore good luck proving with closed models that your question that’s never been asked in any form or derivation (supposedly) is not in the training data.
cristiancavalli··on Modern-Day Oracles or Bullshit Machines? How to thrive in a ChatGPT world
No I can “prove” it — look at any number of cases where LLMs can’t even do basic value comparisons despite being claimed as super intelligent. You can try and say well that’s a limitation of the technology and then I would reply — yes and that’s why I would say it’s not reasoning according the original human definition. Also you have yet to produce any evidence of reasoning and claiming you can over and over again doesn’t add to your arguments substance. I would be interested in your proof that some answer can’t be pattern matched too — at this point I wonder if we could create an non conscious “intelligence” that if large enough would be mostly able to describe anything known to us along some line of probability we couldn’t compute with our brain architecture and it could be close to 99.99999% right. Even if we had this theoretical probability-based super intelligence it still wouldn’t be “reasoning” but could be more “intelligent” than us.

I’m also not entirely convinced we can’t arrive at a reasoning system via probability only (a really cool thought experiment) but these systems do not meet the consistency/intelligence bar for me to believe this currently.

cristiancavalli··on Modern-Day Oracles or Bullshit Machines? How to thrive in a ChatGPT world
Yes people are claiming different things yet no definitive proof has been offered given the varying findings. I can cite another 3 papers which agree with my point and you can probably cite just as many if not more supporting yours. I’m arguing against people depicting what is not a forgone conclusion as such. It seems like in people’s rush to confirm their own preconceived notions people forget that, although a theory may be convincing, it may not be true. Evidence in this very thread of a well-known SOTA LLM not being able to tell which is greater between two numbers indicates to me that what is being called “reasoning” is not what humans do. We can make as many excuses we want per the tokenizer or whatever but then forgive me for not buying the super or even general “intelligence” of this software. I still like these tools though, even if I have to constantly vet everything they say as they often tend to just outright lie, or perhaps more accurately: repeat lies in their training data even if you can elicit a factual response on the same topic.
cristiancavalli··on Modern-Day Oracles or Bullshit Machines? How to thrive in a ChatGPT world
I actually just sat down to crack open MITs Theory of Knowledge and it seems promising and free: https://ocw.mit.edu/courses/24-211-theory-of-knowledge-sprin...

This also looks promising:

https://hiw.kuleuven.be/en/study/prospective/OOCP/introducti...

If you wanted something a bit different Wittgenstein’s Tractatus has always made my head spin with possibilities:

https://people.umass.edu/klement/tlp/tlp-hyperlinked.html

cristiancavalli··on Modern-Day Oracles or Bullshit Machines? How to thrive in a ChatGPT world
Saying something again does not provide proof of its actual veracity. Writing it in caps does not make it true despite the increased emphasis. I default to skepticism in the face of unproven assertions: if one can’t prove that they reason then we must accept the possibility that they do not. There are myriad examples of these models failing to “reason” about something that would trivial for a child or any other human (some are even given as examples in this posts other comments). Given this and the lack of concrete proof I currently tend to agree with the Apple researchers conclusion.
cristiancavalli··on Modern-Day Oracles or Bullshit Machines? How to thrive in a ChatGPT world
This is a fun thought experiment and made me reminisce on my Epistemology classes — something I think the current AI conversation would benefit greatly from. I’m super excited about what we’ve created here — less from the practical standpoint and more from a philosophical one where we get to interact with another form of distilled knowledge. It’s really too bad so much is breathless hype and grift because the philosophy student in me just wants to bask in thinking about this different form/medium/distillation of knowledge we now get to interact with. Comments like these help to reinvigorate that love though so thank you!
cristiancavalli··on Modern-Day Oracles or Bullshit Machines? How to thrive in a ChatGPT world
Disagree — proponents of this point still have yet to prove reasoning and other studies agree about “reasoning” being potentially fake/simulated: https://the-decoder.com/apple-ai-researchers-question-openai...

Just claiming a capability does not make it true and we have 0 “proof” of original reasoning that can be proved coming from these models. Especially given the potential cheating in current SOTA benchmarks

cristiancavalli··on Tesla's Cybertruck is outselling almost every other EV in the US
Not true: https://www.caranddriver.com/news/g60385784/bestselling-cars... Model Y is 4th actually and slipping apparently according to October data.
cristiancavalli··on Ilya Sutskever to leave OpenAI
Do you have a source for any of these numbers or is this just your speculation? I haven’t seen any estimates from well-known institutions that reference any of the numbers your are pointing to.
cristiancavalli··on 23% of bachelor's degrees and 43% of master's degrees have a negative ROI
Not sure why your valid point is being downvoted. You also brought a source which most people don’t bother. Kudos!
cristiancavalli··on 23% of bachelor's degrees and 43% of master's degrees have a negative ROI
You may want to try widening your social circles if this is a recurring theme in your life. Making friends as an adult is hard and making intimate connections sometimes even harder. The best thing I did for myself and my dating/personal life was get involved in charity work. You get to give back, meet people and improve your community. Who knows —- you might even fall in love with your community.
cristiancavalli··on 23% of bachelor's degrees and 43% of master's degrees have a negative ROI
Good for you! You should do what makes you happy and revel in the knowledge that you provide many different kinds of value and you yourself extract value from something you like doing in multiple ways. The way I think about it: no one else has to be me and work my job. Since im going to so the work then I will choose how and what I do so long as I can. It took time for me to build the emotional strength to do what I love and not care about what others think but working on that intentionally has brought so many unforeseen, manifold benefits that I wouldn’t change that path for any other opportunity.
cristiancavalli··on Wage suppression – not stagnation – is costing workers $10 an hour (2021)
Actually it doesn’t — the decline in real income is not huge even on the high end and actual site comments on its findings:

“ How would increasing the minimum wage affect the number of people in poverty? By boosting the income of low-wage workers with jobs, a higher minimum wage would lift some families’ income above the poverty threshold and thereby reduce the number of people in poverty. But low-wage workers who lost employment would see their earnings decrease, and in some cases their family income would fall below the poverty threshold. The first effect would tend to be larger than the second, so the number of people in poverty would generally fall.”

Yes there would be job losses but that effect is shown to not be nearly as large as the number of people lifted out of poverty.

cristiancavalli··on I tried generative AI on lots of data and we're not quite there yet
Would anyone have an idea as to why it would drop large portions of text as pointed out in the article? Would this be a context window issue? Overall very interesting analysis, thank you!!! (I also thought the comparison to the authors bird and wanting to invent conditions to be rewarded to be both funny and intriguing).
cristiancavalli··on HBO CEO Confesses to Trolling Critics with Fake Tweets: Rolling Stone
Best comment of the week right here. Thank you
cristiancavalli··on Wage suppression – not stagnation – is costing workers $10 an hour (2021)
> Firstly, minimum wage is politically unpopular in most areas.

You’re so wrong it’s actually astounding: https://www.pewresearch.org/short-reads/2021/04/22/most-amer... https://www.dataforprogress.org/blog/2023/5/24/725-isnt-cutt... https://www.graydc.com/2023/04/21/study-finds-that-23-americ...

> This is why you see lots of Western countries not even bother with them. Again you’re just so wrong: most of Europe has a much better adjusted minimum wage: https://en.wikipedia.org/wiki/List_of_European_Union_member_...

> Secondly, minimum wages are very inefficient as a poverty reduction policy. For every dollar you impose on businesses, a % is going to be eaten by inflation, a % is going to into reduced labor demand, and a % is going to families well above the poverty line (second or third household earners).

Just wrong again:

https://www.rwjf.org/en/insights/our-research/2022/08/explor...

https://www.brookings.edu/articles/higher-regional-minimum-w...

https://thehill.com/changing-america/respect/poverty/3636046...

Even the CBO generally disagrees with your falsehoods: https://www.cbo.gov/publication/55681

Why spread misinformation?

cristiancavalli··on New York Times tech workers to strike this afternoon
Sure you can move the goal posts, but just so we have a record of the actual history:

You proffered a point: “ If Ford wants to invest in these plants or hire more people, they must hire them from the union. There is no competing, alternative union, or mechanism to hire non-union labor into this plant, thus making the UAW a local monopoly on labor supply for those plants.”

Which I proved incorrect because it’s simply not the case nor could it have been based on Michigans own laws (up until this year that is). Michigan was a right to work state up until this year so yes Ford can and does hire non union which doesn’t even speak to its massive 3pp parts maker network which is largely not unionized and acts as another way to employ non union personnel. You can say Ford can’t hire scabs but they do as do others:

https://labornotes.org/2023/09/scabs-deployed-gm-parts-distr...

https://theintercept.com/2023/10/10/uaw-auto-strike-stellant...

You can also say they can’t hire non union workers and yet even in the US only many ford employees are not unionized: “ Ford has about 173,000 employees worldwide, about 86,000 in the U.S. Of the 86,000, about 57,000 are represented by the UAW. About 33% of Ford employees are UAW” https://www.barrons.com/articles/ford-stock-uaw-gm-stellanti...

I’m not responding to your other “points” because they lack any form of citation. I was interested to see if there were actually union requirements to get a Ford job in Michigan and from my research I can’t see that and your claims fail to speak to that question in an substantive way beyond you espousing your opinion. I never made a claim about unions being good or bad — you imply value judgements in your conclusions but those are your own and without citations they are as useful as a hole in the head. You can keep on making “logical” conclusions but they lack any substance because you can’t properly execute a coherent line of argument as you fail to support your previous claims even when given multiple opportunities to present a cogent case. I guess “everybody knows” though is enough for your purposes.

Also your definition of monopoly in this case fails to actually support your point. Here is an actual definition of a monopoly: https://economictimes.indiatimes.com/definition/monopoly

You miss the important “unique” qualifier to the product or services. UAW or any other union does not offer a unique product as labor is a fungible asset. In right to work states there is no singular unique provider of labor and therefore it is not a monopoly unless you can prove that one has to join a union to work at Ford which is not the case.

cristiancavalli··on New York Times tech workers to strike this afternoon
I mean I just did. Your comment is just plain wrong in the face of the cited right to work law for Michigan. Any “requirements” would be this year alone due to the legality of the statute. It’s your turn to present something other than opinions posed as fact, blatant misinformation or in this response of yours an ad populum appeal that leads me to believe you have spent 0 time actually researching this. How do you square the right to work law in Michigan with your previous claim that Michigan ford shops are “all union” by requirement and not by the workers choice to unionize. How would that even be legal to force someone to join a union in right to work state? Hint: it isn't legal and therefore your claim is 100% provably wrong.
cristiancavalli··on New York Times tech workers to strike this afternoon
Which industries? If you’re gonna point to SAG then sure there are union gigs but that does not preclude one from becoming an actor and working in non union gigs. Perhaps you mean non right to work states?
cristiancavalli··on New York Times tech workers to strike this afternoon
You’re gonna have to cite some sources. I’ve been looking and can’t find a single piece of evidence to back up your claim of union only shops in Michigan. To the contrary Michigan was a right to work state until 2023:

“In 2012, Michigan enacted a right-to-work statute that prevented employees from being forced to join or financially support a labor union as a condition of employment” Which means they can’t force you to join the union.

Maybe you’re use of the word “effectively” is meant to carry the weight of this oversight — although that would imply that, since the workers were not forced to join the union they did so of their own volition? Which I’m sure is not the point you are trying to make.

Where’s the evidence of the claims?

cristiancavalli··on New York Times tech workers to strike this afternoon
Hey at least I offered sources and if you want to throw out ~20 years of data that is pretty damn recent on the premise that business today is so different from the 80s then that’s your no true Scotsman to deal with. Funnily enough some of the most successful business minds today came up in the 80s and 90s and in that culture of business so I think you’re fooling yourself if you think it’s so different today. Also I offered many sources to your total lack of, so being disingenuous about the amount of sources I offered in construction of my argument further leads my to believe you have nothing of actual substance to proffer either in fact or opinion.
cristiancavalli··on New York Times tech workers to strike this afternoon
Conceded that sag has a pretty special case here in the “must join” of course, speaking from the free market view, you are more than welcome to work on non-union gigs of which there are plenty. No one is making you join a union just because you are in a profession + a place. There’s plenty of non-union gigs in Hollywood. I don’t see the harm in making people pay dues if they are getting the benefits of a union gig and this is not the same as “being forced” to join a union — you can go work in another state with right to work laws. This also seems like a pretty special case to SAG since we can see union and non union workers in auto plants: evidenced by the tier system that many auto manufacturers put in place. So not all unions have this same bargaining power.
cristiancavalli··on New York Times tech workers to strike this afternoon
You offer no data or citation to evidence your claims so I won’t bother attempting to refute your opinions.

> In large I feel your response missed the point I’m making.

Which is what? Beyond casting aspersions and making claims with no data the only point I see you making is: “I think I can do better by myself” which is an assertion laughably disproved by the basic history of labor organization.

I feel you missed the point when your argument failed to make cogent case as to why unions would actually hurt your job prospects beyond “a feeling.” Come back with some data and maybe you might actually have point that can be argued.

cristiancavalli··on New York Times tech workers to strike this afternoon
> it’s missing the point of a hyper-transactional, individualist economy that’s only getting more so.

This runs counter to the fact that we have seen a massive increase in labor organization and union activity over the last 5 years so I would consider rethinking that perspective.

https://www.employmentlawworldview.com/recent-data-confirms-...

https://news.gallup.com/poll/510281/unions-strengthening.asp...

> it’s not like we’re miners, stuck in the same valley in WV by a variety of conditions, so might as all band together and fight Duke Energy.

Yes but your are subject to an increasingly consolidated group of employers that are all generally described as “near monopolies” by the stock market analysts themselves and as we saw with the credit tightening cycle startups only are a viable counterweight as an employer when credit is cheap. You are also increasingly subject to competing with people who live on and require far less money to live in the places they work outside of western developed nations. Any handwaving about “superior quality from the US” is just whistling past the graveyard as monopolistic corporations don’t have to care about quality. Anecdata: any post on HN complaining about a major corporations support — they don’t have to care about quality when they own the market.

https://www.openmarketsinstitute.org/learn/monopoly-by-the-n...

“The US economy is at least 50% more concentrated today than in 2005, they write.” https://www.axios.com/2022/05/26/new-paper-finds-monopolies-...

> Also I cant imagine how a startup scales if unions get involved.

And yet data to that effect has been collected and found no visible impact on small business: https://www.princeton.edu/~davidlee/wp/unionbf.pdf

cristiancavalli··on New York Times tech workers to strike this afternoon
Can you proffer any evidence/data to that effect? If it is a monopoly how can we have many non-unionized auto workers working for Tesla and Toyota?
← PreviousPage 2 of 4Next →