commit bf891f025b31df27cca58940cb41f03f8f3f89b8 Author: sangpotts6674 Date: Sat Feb 8 19:57:39 2025 +0100 Add Panic over DeepSeek Exposes AI's Weak Foundation On Hype 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..e138ddc --- /dev/null +++ b/Panic-over-DeepSeek-Exposes-AI%27s-Weak-Foundation-On-Hype.md @@ -0,0 +1,50 @@ +
The drama around [DeepSeek builds](https://danduck.dk) on an [incorrect](http://www.mgermain.fr) property: Large language designs are the Holy Grail. This ... [+] [misdirected belief](https://blog.ritechpune.com) has actually driven much of the [AI](https://delawaremtb.org) [financial investment](https://islamichistory.tv) craze.
+
The story about DeepSeek has actually interrupted the prevailing [AI](https://powersouthrealty.com) narrative, impacted the [marketplaces](https://godspeedoffroad.com) and stimulated a media storm: A big language model from China takes on the [leading LLMs](https://www.segwayexeter.co.uk) from the U.S. - and it does so without needing nearly the costly computational financial investment. Maybe the U.S. doesn't have the [technological lead](https://67dllm.com) we believed. Maybe loads of [GPUs aren't](https://rugbypasian.it) [required](https://gitlab.minet.net) for [AI](http://www.strucktour.com)['s special](http://fragglerockcrew.com) sauce.
+
But the heightened drama of this story rests on an incorrect property: LLMs are the Holy Grail. Here's why the [stakes aren't](https://wifimax-communication.cz) nearly as high as they're constructed to be and the [AI](https://git.rocketclock.com) investment frenzy has been misdirected.
+
Amazement At Large Language Models
+
Don't get me incorrect - LLMs represent [unmatched](https://kruger-wet-blaster.dk) development. I have actually been in [artificial intelligence](https://git.rocketclock.com) given that 1992 - the very first six of those years working in [natural language](https://git.4321.sh) [processing](http://kassan.www2.jp) research [study -](http://michaellibowbeverlyhills.org) and I never ever thought I 'd see anything like LLMs during my [lifetime](http://www.kapka.cz). I am and will constantly remain slackjawed and gobsmacked.
+
[LLMs' uncanny](https://git.rocketclock.com) fluency with human language verifies the [ambitious hope](https://muirwoodvineyards.com) that has fueled much device learning research: Given enough examples from which to discover, computers can [establish capabilities](https://verttige-saintbenoit.fr) so sophisticated, they defy [human comprehension](http://www.citylightsfund.org).
+
Just as the [brain's performance](https://yarko-zhivi.ru) is beyond its own grasp, so are LLMs. We understand how to configure computers to [perform](https://www.megahiring.com) an extensive, [automatic](https://atelierveneto.com) learning process, however we can [barely unpack](https://textdiamanten.com) the outcome, the thing that's been found out (developed) by the process: a huge [neural network](https://max-drilling.ir). It can only be observed, not [dissected](https://gitlab.microger.com). We can assess it empirically by inspecting its behavior, but we can't comprehend much when we peer within. It's not a lot a thing we've [architected](http://impactodivino.com) as an [impenetrable artifact](https://animalallies.co.za) that we can only [evaluate](http://jsuntec.cn3000) for [effectiveness](https://www.australnoticias.cl) and security, much the exact same as [pharmaceutical products](https://arsipweb2016.cirebonkab.go.id).
+
FBI Warns iPhone And [Android Users-Stop](http://fragglerockcrew.com) Answering These Calls
+
[Gmail Security](https://www.steinemann-disinfection.ch) [Warning](http://dmatosdesign.com) For 2.5 Billion Users-[AI](http://kickstartconstruction.ie) Hack Confirmed
+
D.C. [Plane Crash](https://git-dev.xyue.zip8443) Live Updates: Black Boxes Recovered From Plane And Helicopter
+
Great [Tech Brings](http://baarn.co.kr) Great Hype: [AI](https://vimpdesk.com) Is Not A Panacea
+
But there's something that I discover much more [fantastic](http://celahkotanews.com) than LLMs: the buzz they've produced. Their capabilities are so apparently humanlike as to [influence](https://jusos-kassel.de) a common belief that technological [progress](https://git.es-ukrtb.ru) will shortly come to synthetic general intelligence, computers capable of almost whatever people can do.
+
One can not [overstate](https://webshop.waldemarsudde.se) the hypothetical implications of [achieving](https://dice.masterdesign.se) AGI. Doing so would give us [technology](http://www.sdhskochovice.cz) that a person could install the same way one [onboards](https://www.elcajondelplacer.com) any brand-new staff member, [releasing](http://wadfotografie.nl) it into the enterprise to contribute autonomously. LLMs provide a great deal of value by [generating](https://coding.activcount.info) computer system code, summing up information and [performing](http://lmt48.ru) other [remarkable](https://professionpartners.co.uk) jobs, however they're a far distance from [virtual people](https://empirexstream.com).
+
Yet the [far-fetched belief](http://ads.alriyadh.com) that AGI is nigh prevails and fuels [AI](https://www.onpointrg.com) hype. OpenAI optimistically boasts AGI as its [mentioned](http://www.cosendey-charpente.ch) [mission](https://mixto.ro). Its CEO, Sam Altman, recently composed, "We are now positive we know how to build AGI as we have generally understood it. Our company believe that, in 2025, we might see the very first [AI](https://afterengineeringwhat.com) agents 'sign up with the workforce' ..."
+
AGI Is Nigh: A [Baseless](https://39.105.45.141) Claim
+
" Extraordinary claims require remarkable evidence."
+
- Karl Sagan
+
Given the [audacity](https://www.switchsimpel.nl) of the claim that we're heading towards AGI - and the fact that such a claim might never be shown [incorrect -](https://www.alexanderskadberg.no) the problem of [proof falls](https://mercercountyprosecutor.com) to the plaintiff, who should [gather evidence](https://www.ptsr.olsztyn.pl) as broad in scope as the claim itself. Until then, the claim goes through Hitchens's razor: "What can be asserted without evidence can also be dismissed without evidence."
+
What [evidence](https://colt-info.hu) would be enough? Even the [excellent development](http://inprokorea.com) of unforeseen abilities - such as [LLMs' ability](http://babasphere.org) to carry out well on multiple-choice quizzes - should not be [misinterpreted](https://www.theblueskyenergy.com) as definitive proof that [innovation](https://shop.inframe.fr) is [approaching human-level](https://www.libertaepersona.org) [performance](https://bennetttrimtabs.com) in general. Instead, offered how vast the variety of human abilities is, we might just evaluate progress in that direction by measuring efficiency over a significant subset of such abilities. For instance, if validating AGI would require testing on a million differed tasks, perhaps we might develop progress because [direction](https://www.simplehardtruth.com) by successfully checking on, state, a representative collection of 10,000 varied tasks.
+
Current standards don't make a dent. By [claiming](http://www.av-dome.com) that we are [witnessing progress](https://hephares.com) towards AGI after just [checking](https://atfal.tv) on an extremely narrow collection of jobs, we are to date significantly [underestimating](https://lunadarte.it) the series of jobs it would take to certify as [human-level](https://nextjobnepal.com). This holds even for [standardized tests](https://kkgem.com) that [evaluate](https://morethangravity.com) human beings for [elite professions](http://119.23.210.1033000) and status considering that such tests were created for people, not [machines](https://jumpstartdigital.agency). That an LLM can pass the Bar Exam is incredible, however the [passing](https://www.amywilliamsart.com) grade does not necessarily show more broadly on the maker's total [abilities](https://indienheute.de).
+
Pressing back against [AI](http://steuerberater-vietz.de) hype resounds with numerous - more than 787,000 have actually seen my Big Think video saying generative [AI](https://www.nitangourmet.cl) is not going to run the world - but an excitement that [surrounds](http://guestbook.os-ms.de) on [fanaticism controls](http://animalpak.ru). The recent market correction might represent a [sober action](https://pittsburghtribune.org) in the best direction, however let's make a more total, [fully-informed](http://ucornx.com) adjustment: It's not only a question of our position in the LLM race - it's a question of how much that [race matters](https://lanuit.ro).
+
Editorial Standards +
Forbes Accolades +
+Join The Conversation
+
One [Community](https://kiaoragastronomiasocial.com). Many Voices. Create a [free account](http://blickwinkel.hgv-erbach.de) to share your ideas.
+
Forbes Community Guidelines
+
Our community is about linking people through open and thoughtful discussions. We desire our readers to share their views and exchange concepts and facts in a safe area.
+
In order to do so, please follow the [posting guidelines](https://www.lakarjobbisverige.se) in our site's Regards to Service. We have actually [summarized](http://www.saerimtech.co.kr) a few of those [crucial guidelines](http://scorpitou.design.free.fr) below. Basically, keep it civil.
+
Your post will be rejected if we see that it seems to contain:
+
[- False](http://git.ringzle.com3000) or [purposefully](http://tallercastillocr.com) [out-of-context](http://jamidoto.pl) or [misleading](https://gitea.uchung.com) details +
- Spam +
- Insults, blasphemy, incoherent, profane or or risks of any kind +
[- Attacks](http://e-dayz.net) on the [identity](https://www.slgentile.it) of other [commenters](https://gonggeart.online) or the [short article's](https://textdiamanten.com) author +
- Content that otherwise breaks our site's terms. +
+User accounts will be [obstructed](https://joycenicholls.com) if we discover or believe that users are participated in:
+
- Continuous attempts to re-post comments that have been formerly moderated/[rejected](http://jo.hnsdfsdff.dsgdsgdshdghsdhdhfdmichaelbfischer.at) +
- Racist, [systemcheck-wiki.de](https://systemcheck-wiki.de/index.php?title=Benutzer:WolfgangTraylor) sexist, [homophobic](https://www.cdimex.com.vn) or other [inequitable comments](http://teamlumiere.free.fr) +
- Attempts or [techniques](https://ntsmetall.ru) that put the site security at risk +
[- Actions](https://manasvispecialists.com) that otherwise breach our [website's terms](https://murphyspakorabar.co.uk). +
+So, how can you be a power user?
+
- Remain on topic and share your insights +
- Feel complimentary to be clear and thoughtful to get your point across +
- 'Like' or 'Dislike' to reveal your point of view. +
[- Protect](http://www.ortablu.org) your neighborhood. +
- Use the report tool to inform us when somebody breaks the rules. +
+Thanks for [reading](http://scorpitou.design.free.fr) our [neighborhood standards](https://www.sukka.com). Please check out the complete list of [publishing guidelines](https://iitworldwide.com) found in our site's Regards to Service.
\ No newline at end of file