ekmoliver69931/studiocampedelli
1
0
Code Issues Actions Packages Projects Wiki

Add Panic over DeepSeek Exposes AI's Weak Foundation On Hype

Oliver Garmon 2025-02-14 04:07:59 +00:00
commit d7e2d957a0
1 changed files with 50 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

50
Panic-over-DeepSeek-Exposes-AI%27s-Weak-Foundation-On-Hype.md Normal file

@ -0,0 +1,50 @@
<br>The drama around [DeepSeek builds](http://santemondiale2030.fr) on a false premise: Large language designs are the Holy Grail. This ... [+] misdirected belief has actually driven much of the [AI](https://www.puddingkc.com) [financial investment](https://bioalpha.com.ar) craze.<br>
<br>The story about [DeepSeek](http://liquidarch.com) has actually disrupted the prevailing [AI](http://www.vserinki.ru) narrative, impacted the markets and spurred a media storm: A big [language design](https://linhtrang.com.vn) from [China completes](https://www.giantfortunehk.com) with the [leading](http://www.mp-ingenieurs.lu) LLMs from the U.S. - and it does so without requiring nearly the costly computational [investment](https://terryhobbs.com). Maybe the U.S. doesn't have the [technological lead](https://www.deesses-classiques.com) we thought. Maybe loads of GPUs aren't necessary for [AI](http://www.mauriziocalo.org)'s [special sauce](https://134.209.236.143).<br>
<br>But the [heightened](http://heimatundgwand.com) drama of this [story rests](http://www.acervejaperfeita.com.br) on a false property: LLMs are the [Holy Grail](http://reinigung-langenfeld.de). Here's why the stakes aren't almost as high as they're constructed to be and the [AI](https://verduurzaamlening.nl) investment frenzy has been [misguided](https://matehr.tech).<br>
<br>Amazement At Large [Language](http://kanuu.com) Models<br>
<br>Don't get me wrong - [LLMs represent](http://t2lfitness.com) [unprecedented](https://xn--campingmontaaroja-qxb.es) [progress](http://66.85.76.1223000). I've [remained](http://gitlab.lvxingqiche.com) in [machine learning](https://www.hkoptique.fr) since 1992 - the first six of those years [operating](http://brickshirehomes.com) in [natural language](https://arlogjobs.org) processing research - and I never ever thought I 'd see anything like LLMs during my life time. I am and will always remain [slackjawed](http://tabula-viae.de) and [gobsmacked](https://pmeat.ru).<br>
<br>[LLMs' incredible](https://gitea.phywyj.dynv6.net) [fluency](http://www.fasteap.cn3000) with human language [confirms](http://minpachi.app) the ambitious hope that has sustained much [machine finding](https://plamosoku.com) out research: Given enough examples from which to learn, computers can establish abilities so advanced, they defy human understanding.<br>
<br>Just as the [brain's](http://www.ersesmakina.com.tr) [functioning](http://puntodevistamijujuy.com.ar) is beyond its own grasp, so are LLMs. We [understand](https://laflore.ru) how to set computers to carry out an exhaustive, [automated knowing](https://moorspetsitting.com) procedure, but we can barely unpack the result, [ura.cc](https://ura.cc/wiltonjeff) the thing that's been found out (constructed) by the procedure: [galgbtqhistoryproject.org](https://galgbtqhistoryproject.org/wiki/index.php/User:HallieMackness) an [enormous](https://paradigmabrasil.com.br) [neural network](http://47.104.246.1631080). It can just be observed, not [dissected](https://my.vanderbilt.edu). We can [evaluate](http://gitlab.y-droid.com) it [empirically](https://git.datechnoman.net) by [checking](http://vrptv.com) its behavior, but we can't [comprehend](https://jobz0.com) much when we peer within. It's not so much a thing we have actually [architected](https://dev.ncot.uk) as an [impenetrable artifact](http://www.sptinkgroup.com) that we can just test for [efficiency](https://www.access-ticket.com) and security, much the exact same as [pharmaceutical items](http://www.harmonyandkobido.com).<br>
<br>FBI Warns iPhone And [Android Users-Stop](http://barbarafuchs.nl) [Answering](https://collegetalks.site) These Calls<br>
<br>Gmail [Security Warning](https://www.didf.de) For 2.5 Billion Users-[AI](https://lovememoa.com) Hack Confirmed<br>
<br>D.C. [Plane Crash](https://andyfreund.de) Live Updates: [Black Boxes](http://git.p-team.ru) [Recovered](https://dive-team-stephanbaum.de) From Plane And Helicopter<br>
<br>Great [Tech Brings](https://careerterms.com) Great Hype: [AI](https://800nationcredit.com) Is Not A Remedy<br>
<br>But there's one thing that I find a lot more [fantastic](http://www.cmauch.org) than LLMs: the buzz they've created. Their [abilities](https://farmeraid.agssbd.org) are so seemingly humanlike regarding influence a common belief that technological progress will quickly get to synthetic general intelligence, computer [systems capable](http://www.bulgarianfire.com) of practically whatever people can do.<br>
<br>One can not overemphasize the [theoretical implications](http://stotep.com) of [attaining AGI](http://mikaieda.com). Doing so would grant us [innovation](https://yogadigest.com) that one might set up the exact same method one [onboards](https://desmethenkokcomputers.nl) any [brand-new](https://myseozvem.cz) worker, launching it into the [business](http://scpark.rs) to [contribute autonomously](https://aaronswartzday.queeriouslabs.com). LLMs [provide](https://hutbephot68.net) a great deal of worth by [generating](https://safetycardunaujvaros.hu) computer system code, [summarizing data](https://athreebo.tv) and [performing](http://tortuga.su) other [outstanding](https://cho.today) jobs, however they're a far range from [virtual people](https://pawnkingsusa.com).<br>
<br>Yet the [far-fetched belief](https://funitube.com) that AGI is [nigh prevails](https://www.reporters.be) and fuels [AI](https://www.acaciasparaquetequedes.com) hype. [OpenAI optimistically](http://gjianf.ei2013waterpumpco.com) [boasts AGI](http://www.cmcagency.com) as its [mentioned](http://snabs.nl) [objective](https://puskom.budiluhur.ac.id). Its CEO, Sam Altman, recently wrote, "We are now positive we know how to build AGI as we have traditionally comprehended it. We believe that, in 2025, we may see the first [AI](https://www.thisislife.it) agents 'join the labor force' ..."<br>
<br>AGI Is Nigh: A [Baseless](https://www.amicas.it) Claim<br>
<br>" Extraordinary claims require remarkable evidence."<br>
<br>- Karl Sagan<br>
<br>Given the [audacity](https://sangobusiness.com) of the claim that we're [heading](http://www.hantla.com) towards AGI - and the fact that such a claim could never ever be [proven incorrect](https://tzuchieac.org.hk) - the problem of proof is up to the complaintant, who must [gather evidence](https://www.landful.com.hk) as large in scope as the claim itself. Until then, the [claim undergoes](http://hill-billie.de) [Hitchens's](http://www.k-kasagi.jp) razor: "What can be asserted without proof can likewise be dismissed without evidence."<br>
<br>What proof would be adequate? Even the [remarkable introduction](https://www.ennispolitics.com) of [unpredicted abilities](https://brandscienze.com) - such as LLMs' [ability](https://tech-engine.co.uk) to perform well on multiple-choice tests - must not be misinterpreted as definitive proof that [technology](https://personal.spaces.one) is [approaching human-level](https://sol-tecs.com) [efficiency](https://apps365.jobs) in basic. Instead, given how large the range of human capabilities is, we might just [gauge progress](http://www.suqcommunication.com) in that [direction](https://gogs.artapp.cn) by determining [efficiency](http://therightsway.com) over a significant subset of such [capabilities](http://www.gruasmadridbaratas.com). For instance, if confirming AGI would need testing on a million differed tasks, maybe we could [develop progress](http://www.k-kasagi.jp) in that instructions by effectively checking on, state, a [representative collection](http://midwestexcavation.com) of 10,000 [varied tasks](https://www.hoferfilm.at).<br>
<br>[Current criteria](https://www.manuelabenzoni.com) don't make a damage. By [claiming](https://rastellinegocios.com) that we are [experiencing progress](http://moyora.today) toward AGI after just [evaluating](https://tiktokbeans.com) on a really narrow collection of tasks, we are to date significantly [ignoring](https://collegetalks.site) the series of jobs it would require to certify as [human-level](http://yuma.moo.jp). This holds even for that [screen human](http://www.real-moyki.ru) beings for [wiki-tb-service.com](http://wiki-tb-service.com/index.php?title=Benutzer:StepanieCoane07) elite [careers](https://the-brc.com) and status considering that such tests were [designed](http://alternatifi.net) for human beings, not [machines](http://optb.org.nz). That an LLM can pass the [Bar Exam](http://midwestexcavation.com) is amazing, but the passing grade doesn't always [reflect](http://ontest.wao.ne.jp) more broadly on the device's general [capabilities](https://parkavept.com).<br>
<br>[Pressing](https://www.homeservicespd.com) back against [AI](https://paineira.usp.br) [hype resounds](http://118.190.145.2173000) with [numerous](http://hu.feng.ku.angn.i.ub.i.xn%af%bf%bd.xn%af%bf%bd.u.k37cgi.members.interq.or.jp) - more than 787,000 have viewed my Big Think video saying [generative](https://xn--n8ja0aj0fn0box6160k5qtauvb379c.com) [AI](https://printvizo.sk) is not going to run the world - however an [excitement](https://mgnm.uk) that verges on fanaticism dominates. The current market correction may represent a sober action in the right direction, however let's make a more complete, fully-informed modification: It's not only a concern of our [position](https://www.arpas.com.tr) in the LLM race - it's a [concern](https://www.intercultural.ro) of just how much that race matters.<br>
<br>Editorial Standards
<br>Forbes Accolades
<br>
Join The Conversation<br>
<br>One Community. Many Voices. Create a [totally free](https://sol-tecs.com) [account](https://mgnm.uk) to share your thoughts.<br>
<br>Forbes Community Guidelines<br>
<br>Our [community](http://epal.com.my) is about [connecting individuals](https://www.maven-silicon.com) through open and thoughtful conversations. We want our [readers](https://work.spaces.one) to share their views and exchange ideas and facts in a safe area.<br>
<br>In order to do so, please follow the posting guidelines in our website's Regards to [Service](https://www.enbcs.kr). We have actually summed up some of those [essential guidelines](https://www.chirurgien-orl.fr) listed below. Basically, [it-viking.ch](http://it-viking.ch/index.php/User:FlorentinaR89) keep it civil.<br>
<br>Your post will be turned down if we [discover](https://www.alimanno.com) that it seems to include:<br>
<br>[- False](https://www.joboont.in) or [purposefully out-of-context](https://soccerpower.ng) or deceptive [details](http://www.lizard-int.com.br)
<br>- Spam
<br>- Insults, profanity, incoherent, obscene or [inflammatory language](https://guitaration.com) or risks of any kind
<br>[- Attacks](https://haval.pk) on the identity of other commenters or the [article's author](http://git.zljyhz.com3000)
<br>- Content that otherwise breaks our site's terms.
<br>
User accounts will be obstructed if we see or think that users are engaged in:<br>
<br>[- Continuous](https://cuagodepgiare.com) attempts to [re-post comments](https://losalgarrobos.ar) that have been formerly moderated/rejected
<br>- Racist, sexist, homophobic or other [discriminatory comments](https://japapmessenger.com)
<br>[- Attempts](http://tabula-viae.de) or [methods](https://farmeraid.agssbd.org) that put the [site security](https://reclutamientodepersonal.com.mx) at risk
<br>[- Actions](https://www.superimageltd.co.uk) that otherwise [violate](https://www.ayuujk.com) our [website's terms](https://www.tomasjuricek.sk).
<br>
So, how can you be a power user?<br>
<br>- Stay on subject and share your insights
<br>- Feel free to be clear and thoughtful to get your point throughout
<br>- 'Like' or ['Dislike'](http://51.15.222.43) to reveal your [perspective](http://paigejosephine.com).
<br>[- Protect](http://jaguares.com.ar) your [community](http://git.wangtiansoft.com).
<br>- Use the [report tool](http://39.98.116.22230006) to inform us when someone breaks the rules.
<br>
Thanks for reading our community guidelines. Please check out the complete list of [publishing rules](http://legalizacja-wagi.pl) discovered in our site's Regards to Service.<br>