6 ms·
AI is clearly the hottest field right now. Are there any other fields where a high fraction of the high impact papers (perspective or otherwise) are coming fro
by j7ake 3y ago
AI is clearly the hottest field right now.
Are there any other fields where a high fraction of the high impact papers (perspective or otherwise) are coming from industry rather than academia?
- atleastoptimal 3y ago[flagged]
- low_tech_punk 3y agoand artificial intelligence to beget natural intelligence, completing the loop.
- otabdeveloper4 3y agoThere is no such thing as "intelligence".
- slater 3y agoThere is.
- atleastoptimal 3y agoThen why can I type this sentence when a cereal box can't? And please don't give a snarky "because you have hands" type answer.
- coldtea 3y agoBecause the cereal box is not living matter, for starters. Parent didn't say there's no such thing as life.
- atleastoptimal 3y ago~110-120 IQ response
- coldtea 3y agoYes, overqualified response for a ~80 IQ question
- otabdeveloper4 3y ago> Then why can I type this sentence when a cereal box can't? Give it a couple years. IoT WiFi + chatbot + spambot is yesterday's tech.
- soulofmischief 3y agoMaybe there's no intelligence behind this comment, but intelligence is all around you. It's a clearly defined phenomenon.
- Veraticus 3y agoDo you have, like, prior evidence for this statement, or is it just LLM cult stuff?
- coldtea 3y agoIt's current a popular tech, so it must have no realistic limit. Just like the 1969 moon landing meant spaceships for everybody and space colonies in 30 years...
- Al-Khwarizmi 3y agoThat was more an issue of political will and priorities than technological limits, though.
- coldtea 3y agoI don't it would have moved far, even with 10x the budget. There are very real technical and physical and physiological limits preventing this which might be overcome at some point, but not in a few decades through sheer will, or more money at the problem, or because "it's inevitable".
- atleastoptimal 3y agoStrawman. To assume the similarities between space travel in the 60s and AI now only goes as far as "current popular tech" you're either not arguing in good faith or severely short sighted.
- coldtea 3y agoI'm not making an argument by analogy, as much as making a prediction of a similar trajectory and similar dissapointment. That said, there are tons of examples of hyped "current technologies" sold as panaceas that turned into diminishing returns quite fast.
- atleastoptimal 3y agoSeriously though if u think the implications of AI can be boiled to LLM cult hype you are seriously shortsighted. It's weird since u have a public key on your profile you must be someone seriously into tech. Why are so many skilled tech veterans so naively bearish on AI? Is it pride in the primacy of their skills being threatened by artificial intelligence? Is it an inability to gauge general trends outside their field of narrow expertise?
- visarga 3y agoI think both humans and AI's are only as intelligent as their language data, maybe just one bit more than language
- ben_w 3y agoIt might, it might not. Despise my bullish sentiment on AI (I won't be surprised by LLMs being PhD level in everything in 2 years), it is entirely possible that the LLM approach has practical limits that mean it will always be a jack of all trades and master of none, that mastery can only be achieved by specialised AI such as AlphaZero etc. that the LLMs can call out to and which themselves are hard to create in new domains. This could in turn cause another AI winter, even as the current state of the art models are turned into fully open-sourced commodities that can run locally on cellphones.
- atleastoptimal 3y agoWhether it's LLMs some other new groundbreaking architecture the trends and underlying principles are undeniable. To think it's just another hype cycle at this point is silly, and it's crazy how many people look at crypto and AI and think they're remotely similar.
- ben_w 3y ago> the trends and underlying principles are undeniable. To think it's just another hype cycle at this point is silly Saying that is surely a sign of hype, no? > and it's crazy how many people look at crypto and AI and think they're remotely similar. I don't think they're similar, but I absolutely do recognise why they may seem similar. Lots of humans make quick judgements and then anchor on them, and there are many ways to draw lines between almost anything. It's almost like humans are just stochastic pattern matchers… :P
- coldtea 3y agoThe destiny of all natural intelligence is to extinct itself, by damaging the environment, starting an all-out nuclear war, engineer bio weapons, and in many other ways, or just to be wiped out by the first visiting comet ELE, and in the process power starving and killing the substrates of any artificial intelligence it has created too...
- seydor 3y agoFinance? Mining/oil? Everything engineering with lots of money is bound to drive researchers to the private sector
- SiempreViernes 3y agoHush, we're pretending those deep learning papers are science papers, not engineering reports!
- j7ake 3y agoBut there are almost zero papers in glamour journals (high impact journals) from finance or mining/oil.
- a_bonobo 3y agoA lot of the interesting papers in drug discovery come from industry, clinical trials are not cheap, and much of the data you need for the early discovery phase is hidden away in corporate vaults
- eru 3y agoCryptography has a lot of interesting stuff coming from both industry and academia. Especially zero knowledge proofs.