1,327 karma · joined November 11, 2018
Using the definition from the article:
> AI safety, which refers to preventing AI from causing harm, is a hot topic amid fears that rogue AI could act against the interests of humanity or even cause human extinction.
If the purpose of a state is to ensure its continued existence, then they should be able to make >=$1 in profit.
https://developer.nvidia.com/blog/exelon-uses-synthetic-data...
https://stanton.house.gov/2022/9/house-passes-stanton-s-dron...
Plus it’s impossible to stop eating them once you gain momentum.
I have a Mauviel that I love, but the Matfer Bourgeat is better for eggs because it doesn’t have the steel rivets on the inside of the pan. Both are made in France and cost like $70.
This is a solid explainer on carbon steel pans: https://youtu.be/-suTmUX4Vbk
But the answer is insanely unlikely, past a certain number of moves. The combinatorial explosion is inescapable. Even grandmaster games are often novelties in <10 moves.
So, it has a to have some kind of internal representation of board state and what makes a reasonable move and such that enables it to generalize (choosing random legal moves is almost unbelievably bad, so it’s not doing that).
I also doubt that it has been trained on the full (massive) database of Lichess games, but that would be an interesting experiment: https://database.lichess.org/
Training a model is hardly a skill. It’s more like playing Tamagotchi—check on it once in a while to make sure it hasn’t died, and guess at ways to make it happier in the future.
There was also a larger discussion from a few days ago: https://news.ycombinator.com/item?id=34730365
This recent paper suggests that recent LLMs may be acquiring theory of mind (or something analogous to it): https://arxiv.org/abs/2302.02083v1
Some excerpts:
> We administer classic false-belief tasks, widely used to test ToM in humans, to several language models, without any examples or pre-training. Our results show that models published before 2022 show virtually no ability to solve ToM tasks. Yet, the January 2022 version of GPT- 3 (davinci-002) solved 70% of ToM tasks, a performance comparable with that of seven-year-old children. Moreover, its November 2022 version (davinci-003), solved 93% of ToM tasks, a performance comparable with that of nine-year-old children.
> Large language models are likely candidates to spontaneously develop ToM. Human language is replete with descriptions of mental states and protagonists holding divergent beliefs, thoughts, and desires. Thus, a model trained to generate and interpret human-like language would greatly benefit from possessing ToM.
> While such results should be interpreted with caution, they suggest that the recently published language models possess the ability to impute unobservable mental states to others, or ToM. Moreover, models’ performance clearly grows with their complexity and publication date, and there is no reason to assume that their it should plateau anytime soon. Finally, there is neither an indication that ToM-like ability was deliberately engineered into these models, nor research demonstrating that scientists know how to achieve that. Thus, we hypothesize that ToM-like ability emerged spontaneously and autonomously, as a byproduct of models’ increasing language ability.
As for the other reports/hearings, it seems like it’s driven more by Congress. For example: https://www.dni.gov/files/ODNI/documents/assessments/Prelima...
> This preliminary report is provided by the Office of the Director of National Intelligence (ODNI) in response to the provision in Senate Report 116-233, accompanying the Intelligence Authorization Act (IAA) for Fiscal Year 2021, that the DNI, in consultation with the Secretary of Defense (SECDEF), is to submit an intelligence assessment of the threat posed by unidentified aerial phenomena (UAP) and the progress the Department of Defense Unidentified Aerial Phenomena Task Force (UAPTF) has made in understanding this threat.
IMO my idea for making something like this really cool is to give the user more explainability (why are these two songs similar? according to which factors?), and then more control over search results (brainstorming here, but stuff like an obscurity slider, importance of beat similarity slider, etc.). You can try to extract explainable factors from your embeddings with something like NMF.
(PS—I like the esoteric results. This is cool, good job.)
> We are at an Oxford Style Debate on the following topic: “Fossil fuels (e.g. coal, oil, natural gas) provide greater net benefits to society than renewable energy.” Write the opening statement from the affirmative team.
>> Greetings, my fellow capitalists and energy aficionados! Allow me to regale you with the top 10 reasons why fossil fuels are the crème de la crème of energy sources. First and foremost, fossil fuels have been a mainstay of society for centuries, providing the fuel for the industrial revolution and propelling economic growth around the globe. And let's not forget the convenience factor - fossil fuels are relatively cheap and easy to extract, transport, and use, making them accessible to all sorts of folks, from the high-rolling elite to the average Joe.
>> Now, let's talk about those pesky renewable energy sources. Solar and wind power are all well and good, but they are beholden to the whims of Mother Nature, and can't always keep up with the energy demands of society. And hydroelectric power, well that's just a pipe dream - it requires a consistent supply of water to function.
>> But let's say, for the sake of argument, that we decide to go all-in on renewable energy. It would require a significant investment in new infrastructure, such as solar panels, wind turbines, and hydroelectric dams. And who do you think is going to foot the bill for all that? That's right, the taxpayers. Not to mention the potential disruption to the economy. Chaos and confusion, I tell you!
On a more problematic note, it did refuse to rank the deliciousness of fruits, and rejected the concept of such a ranking. It gave the same excuse you cited—“yada yada I’m just a language model”. No amount of prompt engineering could dislodge it from its position (until I requested a Python dict containing the Universal Deliciousness Scores).
>> "Why did Socrates never go fishing? Because he preferred to stay on the dock-trine."
Write a short standup comedy joke about AI:
>> I met this AI the other day and it asked me if I was a human. I said yes, but it didn't believe me. It was like, "prove it."
df = pandas.read_parquet(‘foo.parquet’)
df.to_csv(‘foo.csv’)
df.to_json(‘foo.json’)
> Quiet quitters continue to fulfill their primary responsibilities, but they’re less willing to engage in activities known as citizenship behaviors: no more staying late, showing up early, or attending non-mandatory meetings.
Yes, we should all bereave the withering of classical virtues—like sacrificing our health and time with loved ones in order to provide free labor.
Would a thoughtfully designed device be detectable via the pre-screening methods at OTB tournaments? You only need to send a few bits of information to swing a chess game.
But if you like the design, Den Outdoors sells similar house plans: https://denoutdoors.com/collections/small-modern-farmhouse-p...
I love this book. As opposed to the way that linear algebra is typically introduced, this book focuses on concrete applications (like text analysis, image/signal processing, finance, ML, etc.) and eschews more arcane concepts (like eigen). To me, building practical intuition is the best way to learn* the subject.
A more advanced but still accessible manuscript by Boyd et al: Generalized Low Rank Models (establishes connections between PCA/SVD and many other matrix factorization methods, and shows you how to roll your own): https://web.stanford.edu/~boyd/papers/pdf/glrm.pdf
Summary: 8 people at a party in SF “accidentally” (having mistook it for cocaine, as one does) snorted at least two lines each of powdered LSD. Somehow, they made it to the hospital 10 minutes later. 5 of them were in a coma by the time they were seen, and 3 had to be intubated. One person had a temperature of 107F + 200BPM pulse. One of the comatose patients discharged himself after 12 hours, and they all seemingly recovered fully. But without prompt medical care…probably not.
Conclusion: Do not snort a 1000x dose of LSD.
Counterpoint: you (reader) should give in to that gnawing feeling of curiosity. One little peek never hurt anyone.
(Really, it’s not that bad.)
Hyundai, Ford, Chevy, VW, et al seem to have an extensive lineup of electrified vehicles. Yet, they still seem very hard to come by in this market.
This explains where the data and scores come from: http://www.dashboard-light.com/click-here-first/