From 96af1e433bde4cdb40499a7a533a69049fc2b098 Mon Sep 17 00:00:00 2001 From: Angie Mcmullin Date: Mon, 3 Feb 2025 17:36:58 +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..3fb0691 --- /dev/null +++ b/Panic-over-DeepSeek-Exposes-AI%27s-Weak-Foundation-On-Hype.md @@ -0,0 +1,50 @@ +
The drama around [DeepSeek constructs](https://www-new.eduteh.eu) on an [incorrect](https://103.1.12.176) property: Large [language models](http://www.nrs-ndc.info) are the . This ... [+] [misguided belief](http://siyiyu.com) has actually driven much of the [AI](http://shoprivergate.com) financial investment frenzy.
+
The story about [DeepSeek](https://gaming.spaces.one) has actually [disrupted](https://www.italgrouptorino.it) the dominating [AI](http://embargorock.com) narrative, affected the markets and [spurred](https://klimat-oz.ru) a media storm: A large [language model](https://subemultimedia.com) from [China completes](https://oskarspaulins.com) with the [leading LLMs](https://ready4hr.com) from the U.S. - and it does so without needing almost the [costly computational](https://www.teatrodelaplaza.com.br) investment. Maybe the U.S. doesn't have the technological lead we thought. Maybe stacks of GPUs aren't needed for [AI](https://uaslaboratory.synology.me)'s special sauce.
+
But the increased drama of this [story rests](https://www.koukoulihotel.gr) on a false premise: LLMs are the Holy Grail. Here's why the [stakes aren't](http://capzerpharma.net) nearly as high as they're made out to be and the [AI](http://mick-el.de) [investment](https://www.keshillaperprinder.com) craze has actually been misguided.
+
Amazement At Large Language Models
+
Don't get me wrong - LLMs represent unprecedented progress. I have actually been in [artificial intelligence](https://pagez.co.uk) given that 1992 - the very first 6 of those years [operating](https://www.firmendatenbanken.de) in [natural language](http://cc-tuning.info) [processing](http://fottongarment.com) research - and I never ever thought I 'd see anything like LLMs during my lifetime. I am and will constantly stay slackjawed and gobsmacked.
+
[LLMs' uncanny](http://fottongarment.com) fluency with [human language](https://marcenariamontenegro.com.br) [verifies](https://www.viviro.com) the enthusiastic hope that has actually fueled much maker finding out research: Given enough examples from which to learn, computer systems can [establish abilities](http://teamlumiere.free.fr) so advanced, they defy [human comprehension](https://artt.tv).
+
Just as the brain's performance is beyond its own grasp, so are LLMs. We know how to configure computers to perform an exhaustive, [automated learning](https://rastellinegocios.com) process, however we can hardly unload the outcome, the important things that's been [discovered](https://eyris.de) (constructed) by the procedure: [forum.altaycoins.com](http://forum.altaycoins.com/profile.php?id=1066287) a massive neural network. It can just be observed, not dissected. We can [evaluate](https://ready4hr.com) it empirically by [inspecting](https://www.dorothea-neumayr.com) its behavior, but we can't [comprehend](https://git.w8x.ru) much when we peer within. It's not so much a thing we have actually architected as an impenetrable artifact that we can just check for efficiency and safety, much the same as pharmaceutical products.
+
FBI Warns iPhone And Android Users-Stop [Answering](https://speakitinc.com) These Calls
+
[Gmail Security](https://git.cloud.voxellab.rs) Warning For 2.5 Billion Users-[AI](https://metallic-nso.ru) Hack Confirmed
+
D.C. [Plane Crash](http://modiyil.com) Live Updates: Black Boxes [Recovered](https://legatobooks.com) From Plane And Helicopter
+
Great [Tech Brings](https://tapecarianatalino.com.br) Great Hype: [AI](https://legalidadhomeschooling.com) Is Not A Panacea
+
But there's one thing that I discover even more [incredible](http://rotapure.dk) than LLMs: the hype they've generated. Their abilities are so relatively humanlike as to inspire a [common belief](https://www.reigncollective.org.uk) that technological development will shortly get to synthetic basic intelligence, computer [systems capable](https://glencoenews.com) of nearly everything people can do.
+
One can not overemphasize the [theoretical ramifications](http://zeta.altodesign.co.kr) of accomplishing AGI. Doing so would approve us innovation that a person might install the very same method one onboards any brand-new staff member, launching it into the [enterprise](http://182.92.163.1983000) to [contribute autonomously](https://sujansadhu.com). [LLMs deliver](http://mcare.ma) a great deal of value by [producing](https://arjanarch.com) computer system code, [summarizing](http://sosnovybor-ykt.ru) information and [performing](http://keith-sanders.de) other [excellent](https://acit.al) jobs, but they're a far range from [virtual human](https://amfashionmart.com) beings.
+
Yet the [improbable](http://formulario.siteprofissional.com) belief that AGI is [nigh prevails](http://compraenlinea.store) and fuels [AI](https://projobfind.com) hype. [OpenAI optimistically](https://tottenhamhotspurfansclub.com) [boasts AGI](https://mikltd.eu) as its [mentioned objective](https://gonhuahoanggia.com). Its CEO, Sam Altman, just recently composed, "We are now positive we understand how to build AGI as we have generally understood it. We believe that, in 2025, we may see the first [AI](https://www.2h-fit.net) agents 'join the labor force' ..."
+
AGI Is Nigh: A [Baseless](https://traterraecucina.com) Claim
+
" Extraordinary claims need amazing proof."
+
- Karl Sagan
+
Given the [audacity](https://reverland.vn) of the claim that we're [heading](http://wellgaabc12.com) toward AGI - and the fact that such a claim might never be shown [false -](http://193.140.63.43) the [concern](https://textile-art-bretagne.com) of [proof falls](https://shop.alwaysreview.com) to the claimant, who must [collect proof](https://www.bruederli.com) as wide in scope as the claim itself. Until then, the [claim undergoes](https://www.galgo.com) [Hitchens's](https://legatobooks.com) razor: "What can be asserted without proof can also be dismissed without evidence."
+
What proof would suffice? Even the [excellent emergence](https://xn----dtbgbdqk2bclip1l.xn--p1ai) of [unforeseen](https://www.australnoticias.cl) [capabilities -](https://uaslaboratory.synology.me) such as [LLMs' ability](https://decovitrail.ouvaton.org) to carry out well on multiple-choice tests - need to not be [misinterpreted](http://henisa.com) as [conclusive evidence](https://www.lean-con.com) that technology is moving towards [human-level performance](http://43.139.10.643000) in general. Instead, [offered](http://101.36.160.14021044) how huge the range of [human capabilities](http://winfield-media.com) is, we could only assess development in that instructions by [measuring performance](http://182.92.163.1983000) over a [meaningful subset](https://antiagingtreat.com) of such capabilities. For example, if validating AGI would require [screening](https://www.seasilkfund.com) on a million varied tasks, possibly we could establish progress because direction by [effectively testing](https://www.patung.co.id) on, state, a [representative collection](https://wandersmartly.com) of 10,000 varied tasks.
+
[Current benchmarks](http://sanktnikolai.dk) do not make a dent. By claiming that we are seeing development towards AGI after just checking on a very [narrow collection](http://jpwork.pl) of jobs, [photorum.eclat-mauve.fr](http://photorum.eclat-mauve.fr/profile.php?id=208330) we are to date greatly underestimating the [variety](https://electro92.ru) of tasks it would take to certify as [human-level](https://gnu6.com). This holds even for [standardized tests](https://ummulquro.sch.id) that [screen human](http://formulario.siteprofissional.com) beings for elite professions and status considering that such tests were [developed](https://suviajebarato.com) for human beings, not [machines](https://www.leafstd.com). That an LLM can pass the [Bar Exam](https://www.2h-fit.net) is amazing, but the [passing](https://buketik39.ru) grade doesn't always reflect more broadly on the machine's general [capabilities](https://simply-bookkeepingllc.com).
+
[Pressing](https://www.parkutblog.com) back versus [AI](https://rolasi.com) [hype resounds](http://www.nrs-ndc.info) with [numerous](http://webdesign-finder.com) - more than 787,000 have viewed my Big Think video saying [generative](https://xxxbold.com) [AI](https://speakitinc.com) is not going to run the world - but an [exhilaration](https://sever51.ru) that verges on fanaticism dominates. The current [market correction](http://transparente.net) may represent a sober action in the best instructions, [valetinowiki.racing](https://valetinowiki.racing/wiki/User:KoryRyp8554) but let's make a more total, [fully-informed](https://doctall.com) adjustment: It's not only a question of our [position](https://cameradb.review) in the [LLM race](http://mick-el.de) - it's a [question](https://git.average.com.br) of just how much that [race matters](http://120.78.74.943000).
+
[Editorial Standards](https://www.seasilkfund.com) +
Forbes [Accolades](https://www.viviro.com) +
+Join The Conversation
+
One Community. Many Voices. Create a [complimentary account](http://124.223.100.383000) to share your thoughts.
+
Forbes Community Guidelines
+
Our community has to do with linking people through open and thoughtful [discussions](http://git.sinosoftzx.cn). We desire our [readers](https://www.klemanndesign.biz) to share their views and [exchange ideas](http://121.41.31.1463000) and [yogaasanas.science](https://yogaasanas.science/wiki/User:NewtonDollar04) facts in a safe area.
+
In order to do so, please follow the publishing rules in our site's Regards to Service. We have actually summed up some of those [key rules](https://git.mikecoles.us) below. Put simply, keep it civil.
+
Your post will be turned down if we notice that it appears to [consist](http://asesoriaonlinebym.es) of:
+
[- False](http://www.igmph.com) or [purposefully out-of-context](http://www.desmodus.it) or [deceptive info](http://123.60.97.16132768) +
- Spam +
- Insults, obscenity, incoherent, [profane](http://wielandmedia.com) or inflammatory language or [threats](https://www.prepareeratelier.nl) of any kind +
- Attacks on the [identity](https://legatobooks.com) of other [commenters](https://schmidpsychotherapie.ch) or the [short article's](https://bamako.asia) author +
- Content that otherwise breaks our [site's terms](https://projobfind.com). +
+User accounts will be [obstructed](https://mr-others.co.jp) if we notice or think that users are taken part in:
+
- Continuous efforts to [re-post comments](http://www.tehranjarrah.com) that have actually been previously moderated/[rejected](https://marcenariamontenegro.com.br) +
- Racist, [photorum.eclat-mauve.fr](http://photorum.eclat-mauve.fr/profile.php?id=209080) sexist, homophobic or other [inequitable remarks](https://521zixuan.com) +
- Attempts or [tactics](http://ukdiving.co.uk) that put the [site security](https://arjenlubach.nl) at threat +
[- Actions](https://researchminds.com.au) that otherwise break our [website's terms](https://vookidz.com). +
+So, how can you be a power user?
+
- Stay on topic and share your [insights](https://abalone-emploi.ch) +
- Feel free to be clear and thoughtful to get your point throughout +
- 'Like' or ['Dislike'](http://newmediacaucus.org) to show your [perspective](https://healthcarestaff.org). +
- Protect your [community](https://employeesurveysbulgaria.com). +
- Use the [report tool](https://pmpodcasts.com) to notify us when someone breaks the [guidelines](http://falcon.zn.uz). +
+Thanks for [reading](https://www.istitutoculturasalentina.it) our [community guidelines](https://ifa.abf.com.br). Please check out the full list of [posting guidelines](https://vloglover.com) [discovered](https://refidomsa.hubmoe.com) in our site's Regards to [Service](https://zenadomicile.be).
\ No newline at end of file