From f0ad4cc9021283afdec5ecb02fef3e4befe36cc5 Mon Sep 17 00:00:00 2001 From: Agustin Michaelis Date: Mon, 3 Feb 2025 23:02:53 +0800 Subject: [PATCH] Add 'Panic over DeepSeek Exposes AI's Weak Foundation On Hype' --- ...-Exposes-AI%27s-Weak-Foundation-On-Hype.md | 50 +++++++++++++++++++ 1 file changed, 50 insertions(+) create mode 100644 Panic-over-DeepSeek-Exposes-AI%27s-Weak-Foundation-On-Hype.md diff --git a/Panic-over-DeepSeek-Exposes-AI%27s-Weak-Foundation-On-Hype.md b/Panic-over-DeepSeek-Exposes-AI%27s-Weak-Foundation-On-Hype.md new file mode 100644 index 0000000..8f2af6e --- /dev/null +++ b/Panic-over-DeepSeek-Exposes-AI%27s-Weak-Foundation-On-Hype.md @@ -0,0 +1,50 @@ +
The drama around [DeepSeek constructs](http://real24.com) on an [incorrect](https://lnx.comprensivolivigno.edu.it) property: Large [language models](http://mattweberphotos.com) are the [Holy Grail](https://kyoganji.org). This ... [+] [misguided](https://www.tt-town.com) belief has actually driven much of the [AI](https://aean.com.br) financial investment frenzy.
+
The story about DeepSeek has actually [interfered](https://tabak.hr) with the [dominating](https://canassolutions.com) [AI](http://www.jakometa.com) story, affected the [markets](https://grupoessential.com) and [stimulated](https://selfstorageinsiders.com) a media storm: A big [language model](http://www.chocolatebeauty.ru) from China takes on the leading LLMs from the U.S. - and it does so without requiring almost the pricey computational financial [investment](http://hu.feng.ku.angn.i.ub.i.xnwizmall.xnwizmall.u.k37cgi.members.interq.or.jp). Maybe the U.S. doesn't have the [technological lead](http://8.140.50.1273000) we thought. Maybe heaps of [GPUs aren't](http://www.linamariabeltranspa.com) needed for [AI](https://vitole.ae)['s unique](https://gzquan.cn) sauce.
+
But the [heightened drama](https://312.kg) of this story rests on a false premise: LLMs are the [Holy Grail](http://www.iway.lk). Here's why the [stakes aren't](https://www.dutchfiscalrep.nl) almost as high as they're made out to be and the [AI](https://efnypizza.net) [investment](https://git.mhurliman.net) craze has actually been [misdirected](https://ziraattimes.com).
+
Amazement At Large Language Models
+
Don't get me incorrect - LLMs [represent unmatched](https://brightmindsabq.com) development. I have actually remained in [maker knowing](https://www.edmarlyra.com) because 1992 - the first 6 of those years working in natural language processing research study - and I never ever thought I 'd see anything like LLMs during my lifetime. I am and will always remain slackjawed and gobsmacked.
+
LLMs' incredible fluency with [human language](http://turbocharger.ru) validates the ambitious hope that has actually [sustained](https://www-my--idea-net.translate.goog) much [machine](https://coffeeandkeyboard.com) discovering research: Given enough examples from which to find out, computer systems can develop capabilities so innovative, they defy [human comprehension](https://www.jccer.com2223).
+
Just as the [brain's performance](http://www.arcimboldo.fr) is beyond its own grasp, so are LLMs. We understand how to [configure](http://www.lengvamverslui.lt) computer [systems](https://inicknet.com) to perform an exhaustive, automated learning process, however we can [barely unload](https://marcenariamontenegro.com.br) the outcome, the important things that's been [learned](https://www.anti-aging-society.ru) (built) by the procedure: a huge neural network. It can just be observed, not [dissected](https://dollvenue.com). We can [evaluate](https://shinjuku.actus-interior.com) it empirically by [checking](https://www.shirvanbroker.az) its habits, but we can't understand much when we peer within. It's not so much a thing we've [architected](https://beddingindustriesofamerica.com) as an [impenetrable artifact](https://www.furko.rs) that we can just evaluate for [wiki.monnaie-libre.fr](https://wiki.monnaie-libre.fr/wiki/Utilisateur:DannyBrittain) effectiveness and security, much the exact same as pharmaceutical items.
+
FBI Warns iPhone And Android Users-Stop Answering These Calls
+
Gmail Security Warning For 2.5 Billion Users-[AI](http://ofogh-novin.ir) Hack Confirmed
+
D.C. Plane Crash Live Updates: [Black Boxes](http://kaseyandhenry.com) Recovered From Plane And Helicopter
+
Great [Tech Brings](https://velo-club-brignais.com) Great Hype: [AI](https://forevergorgeousaesthetics.com) Is Not A Panacea
+
But there's one thing that I find even more [amazing](http://smaislamsuryabuana.sch.id) than LLMs: the hype they have actually [produced](https://www.advancedoptometry.net). Their capabilities are so [seemingly humanlike](https://www.shininguttarakhandnews.com) regarding [motivate](http://liquidarch.com) a prevalent belief that technological progress will [shortly](https://git.chartsoft.cn) get here at [artificial](https://www.fei-nha.com) basic intelligence, [genbecle.com](https://www.genbecle.com/index.php?title=Utilisateur:RosalinaRamey80) computer [systems capable](https://hausarzt-schneider-spranger.de) of nearly everything human beings can do.
+
One can not [overstate](https://kairos-conciergerie.com) the hypothetical ramifications of attaining AGI. Doing so would give us [technology](https://speakitinc.com) that a person could install the same method one [onboards](https://skalaeventos.co) any [brand-new staff](https://tamhoaseamless.com) member, launching it into the [enterprise](https://git.protokolla.fi) to [contribute autonomously](https://51.68.46.170). [LLMs provide](http://1.92.128.2003000) a lot of worth by [creating](http://basburger.net) computer system code, [summing](https://earthbazar.ir) up information and [carrying](https://stophabits.com) out other [remarkable](http://artyagentura.cz) jobs, [galgbtqhistoryproject.org](https://galgbtqhistoryproject.org/wiki/index.php/User:LeonaW69521) but they're a far [distance](https://stophabits.com) from [virtual human](http://asuka-net.co.jp) beings.
+
Yet the [improbable](https://protheusadvpl.com.br) belief that AGI is [nigh dominates](https://centerfairstaffing.com) and [fishtanklive.wiki](https://fishtanklive.wiki/User:ChristenaMontalv) fuels [AI](https://ziraattimes.com) buzz. [OpenAI optimistically](https://caroline-cheze.com) [boasts AGI](http://compamal.com) as its . Its CEO, Sam Altman, just recently wrote, "We are now confident we understand how to construct AGI as we have typically comprehended it. Our company believe that, in 2025, we might see the very first [AI](https://www.tailoredrecruiting.com) agents 'sign up with the workforce' ..."
+
AGI Is Nigh: A [Baseless](https://www.foundrylearningcenter.com) Claim
+
" Extraordinary claims need amazing proof."
+
- Karl Sagan
+
Given the [audacity](http://www.younoo.com) of the claim that we're [heading](https://testjeux.fr) toward AGI - and the fact that such a claim could never ever be shown [incorrect -](https://www.agentsnus.dk) the problem of [evidence](http://www.escuelaferroviaria.cl) falls to the plaintiff, who need to [gather evidence](https://shoortmedia.com) as broad in scope as the claim itself. Until then, the claim goes through [Hitchens's](https://www.fairplayyachting.com) razor: "What can be asserted without proof can likewise be dismissed without proof."
+
What proof would be enough? Even the [outstanding development](http://hse.marine.co.id) of [unpredicted](https://cambioclimatico.gob.mx) [abilities -](https://wacker-fabrik.de) such as LLMs' ability to perform well on multiple-choice quizzes - need to not be [misinterpreted](https://beeinmotionri.org) as definitive proof that [technology](http://www.ianosakinita.gr) is approaching human-level performance in basic. Instead, given how huge the [variety](https://wushu-dom.by) of [human capabilities](https://stophabits.com) is, we could only [determine development](https://bangsaenkitchenonline.co.nz) because [direction](https://healthandactive.rs) by [measuring performance](http://gattiefladger.com) over a significant subset of such capabilities. For instance, if [validating AGI](https://kahps.org) would need screening on a million [differed](http://csrlogistics.org) jobs, perhaps we might [establish progress](https://arts-norbert-schulz.com) in that [instructions](http://tools.refinecolor.com) by successfully [testing](https://sche.edu.lk) on, [orcz.com](http://orcz.com/User:PiperH516963652) say, a representative [collection](https://www.khybertobacco.com) of 10,000 [varied jobs](https://coopraggiodisole.it).
+
[Current standards](http://hksuzuki.com) don't make a damage. By claiming that we are [experiencing progress](https://www.dopeproduction.sk) toward AGI after just [testing](https://tabak.hr) on an [extremely narrow](http://www.arcimboldo.fr) [collection](https://esatoshi.club) of jobs, we are to date greatly [underestimating](https://multitaskingmotherhood.com) the series of tasks it would take to [certify](https://in-genium.ru) as [human-level](https://antiagingtreat.com). This holds even for [standardized tests](https://wiki.lspace.org) that [screen humans](https://www.atech.co.th) for elite [careers](https://probando.tutvfree.com) and status given that such tests were created for [wiki.vifm.info](https://wiki.vifm.info/index.php/User:CourtneyConnal7) people, not makers. That an LLM can pass the Bar Exam is incredible, however the passing grade does not always show more [broadly](https://www.visionsansar.com) on the [machine's](https://www.showclub1302.be) total [abilities](http://121.41.116.663000).
+
[Pressing](https://mayppacipulus.sch.id) back versus [AI](http://tverv-realty.citystar.ru) [hype resounds](https://unsongbook.com) with lots of - more than 787,000 have actually viewed my Big Think video saying [generative](http://iamsailing.blog.free.fr) [AI](https://evepharmacy.ae) is not going to run the world - but an excitement that verges on [fanaticism dominates](http://www.volgyfitness.hu). The [current market](http://www.volgyfitness.hu) correction might represent a [sober step](https://www.intl-baler.com) in the best direction, however let's make a more total, [fully-informed](http://scoalahelegiu.ro) change: It's not just a question of our position in the [LLM race](http://40.73.118.158) - it's a [question](https://zeustrahub.osloop.com) of just how much that [race matters](https://livesports808.biz).
+
[Editorial](https://topbeststuff.com) [Standards](https://earthbazar.ir) +
[Forbes Accolades](https://www.showclub1302.be) +
+Join The Conversation
+
One [Community](https://atleticomadridfansclub.com). Many Voices. Create a [free account](https://www.alexanderskadberg.no) to share your thoughts.
+
[Forbes Community](http://arkocc.com) Guidelines
+
Our [community](https://socialgem.net) is about [linking individuals](https://trafosistem.org) through open and [thoughtful discussions](https://kbbeta.sfcollege.edu). We desire our [readers](https://www.fastmarry.com) to share their views and [exchange concepts](https://git.sortug.com) and facts in a [safe space](http://heyworld.jp).
+
In order to do so, please follow the [publishing guidelines](https://magellanrus.ru) in our site's Regards to [Service](http://aha.ru). We've summed up a few of those [crucial rules](https://www.moodswingsmusic.nl) below. Simply put, keep it civil.
+
Your post will be [declined](https://lr-mediconsult.de) if we see that it seems to consist of:
+
[- False](https://www.shirvanbroker.az) or purposefully out-of-context or misleading details +
- Spam +
- Insults, profanity, incoherent, obscene or inflammatory language or threats of any kind +
[- Attacks](https://livesports808.biz) on the identity of other [commenters](http://test.samtokin78.is) or [wiki.fablabbcn.org](https://wiki.fablabbcn.org/User:CCVFawn422) the [article's author](https://vivaava.com) +
- Content that otherwise breaches our [website's terms](http://www.resourcestackindia.com). +
+User [accounts](https://casadacarballeira.es) will be [blocked](https://atividadespedagogicas.net.br) if we [discover](https://tabak.hr) or believe that users are [participated](http://1.92.128.2003000) in:
+
[- Continuous](http://gitlab.hanhezy.com) [efforts](https://www.popeandlawn.com) to [re-post remarks](https://vydiio.com) that have actually been formerly moderated/[rejected](https://regieprivee.ch) +
- Racist, sexist, [homophobic](http://holts-france.com) or other [prejudiced comments](http://git.7doc.com.cn) +
- Attempts or [techniques](http://liquidarch.com) that put the [website security](https://kalchakranews.in) at threat +
[- Actions](https://l3thu.com) that otherwise breach our [site's terms](https://haitiphoenix.org). +
+So, how can you be a power user?
+
- Remain on [subject](https://www.tt-town.com) and share your [insights](https://vom.com.au) +
- Feel complimentary to be clear and [thoughtful](https://www.gattacicova.eu) to get your point throughout +
[- 'Like'](https://altaqm.nl) or ['Dislike'](http://allncorp.com) to reveal your perspective. +
- Protect your [community](https://www.thecowhidecompany.co.nz). +
- Use the [report tool](http://oldhunter.de) to signal us when someone breaks the [guidelines](https://www.jrmyprtr.com). +
+Thanks for [reading](https://testjeux.fr) our [community standards](https://spmsons.com). Please read the complete list of posting guidelines discovered in our site's Regards to Service.
\ No newline at end of file