Kwiintlobo ezimbini ezimbini zokusetyenziswa kwihlabathi ezidlulileyo kwihlabathi ezidlulileyo, iziphumo ezininzi zophuhliso zobuchwepheshe ziye zihlanganisa kunye neziphumo ezininzi ezininzi zokusetyenziswa kwihlabathi. ukusuka kwimeko yobugcisa kunye ne-colonial sovereignty ukuba umntu wehlabathi wehlabathi Oh wait, ndingathanda umhlaba   Ngoko ke. Iingxaki ezininzi zehlabathi, Thenga iinkampani ezininzi Ngaba changed that much Thenga iinkampani ezininzi Ngaba Ngaba Ukuguqulwa kakhulu Ukukhangisa ngaphandle, imibuzo ezininzi efana neengxaki zehlabathi malunga neengxaki ezijolisweyo malunga neengxaki ze-humanity. Ngokuvamile, imibuzo ye-climate ibizwa njengomnye yeengxaki ezidlulileyo ezivela abasebenzi, iinkwenkwezi kunye nabazi. Ngokuvamile, ubuchwepheshe zethu ezihlangene ngokukhawuleza, ukusuka kwiminyaka eyi-19 century, babe isiphumo esizayo yokushisa kwehlabathi ngokusebenzisa ukuchithwa kweengxasi ze-greenhouse kunye neengxaki. Nangona kunjalo, i-AI inokufumaneka njenge-exception ukusuka le pattern – ngokutsho, ukusetyenziswa kwe-intelligent energy. Izixhobo ezininzi ezininzi ezininzi ezininzi ezininzi ezininzi ezininzi ezininzi ezininzi ezininzi ezininzi ezininzi ezininzi ezininzi ezininzi ezininzi ezininzi ezininzi ezininzi ezininzi. High-Power Computing Facilities (HPCs) High-Power Computing Facilities (HPCs) Energy Sink, Pollutant Faucet Qaphela: Le nqaku, kunye neqaku elandelayo, iqukethe iinkcukacha ezivela kwi-Berkeley Lab 2024 United States Data Center Energy Usage Report kunye ne-Nature report on the Environmental Impact of Large Language Models. I-Berkeley Lab 2024 I-United States Data Center Ukusetyenziswa kwe-Energy Report I-Berkeley Lab 2024 I-United States Data Center Ukusetyenziswa kwe-Energy Report I-Nature Report on the Environmental Impact of Large Language Models I-Nature Report on the Environmental Impact of Large Language Models Ukubala ngalinye i-energy eyasetyenziswa kwindawo ye-HPC kunokwenzeka. Nangona izixeko ze-server ezivamile, njenge-hosting yewebhu, zihlanganisa i-rack ye-Central Processing Units (CPUs) kakhulu efana ne-computer yakho isebenzisa, izixeko ze-AI ezizodwa zihlanganisa ngezindlela ezininzi ezininzi: 
 
 
 
 
 Izakhiwo ze-HPC ziquka ukusetyenziswa kwe-Graphic Processing Units (GPUs) ngaphandle kwe-Central Processing Units (CPUs), ikakhulu ngenxa yokusebenza okuphakeme kwe-CPU kwi-parallel processing kunye ne-energy efficiency. Izakhiwo ze-HPC zisebenzisa iindlela ze-bandwidth ephezulu kunye ne-transfer ngenxa yeemfuno ezininzi ze-AI data processing. Izakhiwo ze-HPC ziquka iinkqubo ze-cooling ezizodwa ngenxa ye-heat eyenziwe ngama-array ezininzi ze-GPU. Izixhobo ze-HPC zisetyenzise amandla ye-computing ngokufanelekileyo (ukugqibela umsebenzi wokucubungula idatha) ngaphandle kwe-variable (i-500-20,000 abasebenzisi abavela ukufikelela kwiwebhusayithi). Ukusetyenziswa ngempumelelo Kwimeko, lezi zihlanganisa kubalulekile ukuba iziko ze-AI zokusetyenziswa zangaphakathi zangaphakathi zangaphakathi zangaphakathi zangaphakathi zangaphakathi zangaphakathi zangaphakathi zangaphakathi zangaphakathi zangaphakathi zangaphakathi zangaphakathi zangaphakathi zangaphakathi zangaphakathi zangaphakathi zangaphakathi zangaphakathi zangaphakathi zangaphakathi zangaphakathi zangaphakathi zangaphakathi zangaphakathi zangaphakathi zangaphakathi zangaphakathi zangaphakathi zangaphakathi zangaphakathi zangaphakathi zangaphakathi zangaphakathi zangaphakathi zangaphakathi zangaphakathi zangaphakathi zangaphakathi. Ukusetyenziswa kwezinto ezininzi zokusetyenziswa kwinkqubo zokusetyenziswa kwinkqubo zokusetyenziswa kwinkqubo zokusetyenziswa kwinkqubo zokusetyenziswa kwinkqubo zokusetyenziswa kwinkqubo zokusetyenziswa kwinkqubo zokusetyenziswa kwinkqubo zokusetyenziswa kwinkqubo zokusetyenziswa kwinkqubo zokusetyenziswa kwinkqubo zokusetyenziswa kwinkqubo zokusetyenziswa kwinkqubo zokusetyenziswa kwinkqubo zokusetyenziswa kwinkqubo zokusetyenziswa kwinkqubo zokusetyenziswa kwinkqubo zokusetyenziswa kwinkqubo zokusetyenziswa kwinkqubo zokusetyenziswa kwinkqubo zokusetyenziswa kwinkqubo zokusetyenziswa kwinkqubo zokusetyenziswa kwinkqubo zokusetyenz Relative Efficiency? I-AI (ngokusetyenziswa ngokufanelekileyo, ngoko, ngoko) ibonelela kakhulu kumazwe lethu. Nangona iinkampani ezininzi kunye nabasebenzi kwaye ngokugqithisileyo kunye ne-AI, Iingcebisi waya ukucacisa ngexabiso efunyenweyo i-AI yokusabalalisa nathi ngokubala iindleko ezininzi ze-500 iingcebisi ezisetyenzisiweyo ukusetyenziswa kwezixhobo ze-AI kunyango kunye nokuthintela iingcebiso ze-humane kunye nokuthintela ixesha, iingcebiso ze-economic, i-emissions ye-carbon, kunye nokusetshenziswa kwamanzi, iingcebisi bafumane iingcebiso ze-qualitative ye-writing kunye ne-accounting. I-Llama-3 ye-Meta (i-LLMs) ye-Large Language Models (i-LLMs) ezihambelana ne-Llama-3 ye-Meta (i-LLMs) ezihambelana neengxaki ze-40 ukuya kwi-150 ze-Americans, kwaye iimodeli ezincinane kunye ne-energy-efficient ezihambelana ne-Google Gemma-2B zibonakalisa ukusebenza kwizinga ze-130-1100 ze-Americans. reported working faster Ukucinga Ukufumana le Iinkampani ezisebenza ngokukhawuleza Ukufumana le Yintoni oku kuthetha ukuba kuqala ukuguqula abasebenzi abantu nge-AI? Ngokungafani ne-gravity ebandayo ye-workers displacement kwi-global society and economy, ukusetyenziswa kwe-AI, njengoko ithi, ibonakala neengxaki ebandayo yokuba akuyona ngaphandle kokuhlawulelwa. Enye ingxaki ezininzi ebandayo kwimodeli yabasetyenziswa yi-LLM yi-incapacity yokuthintela umgangatho weemveliso eyenziwe ngabantu kunye ne-AI. Ngokungafani oku, sinokufumana ukuba, xa i-AI ibonwe ukwandisa iimveliso yeempahla ze-umntu, ukunyuka kwimveliso ye-AI kunye ne-inventiveness ye-umntu kunokhiqiza umsebenzi ebandayo, i-LLMs ezahlukileyo akakwazi ukuvelisa umsebenzi ebandayo ngaphandle, ngex Nangona kunjalo, emva kokufunda, 
 
 Le nqaku lithunyelwe kwiphepha le-Our AI, i-AI Ethics organization eyenziwe ngama-students kunye ne-students eyenza ukwandisa iinkcukacha kwi-AI ngaphandle kwexesha elidlulileyo e-media. Ukuba ukhangela le nqaku, nqakraza iimpazamo zethu yenyanga kunye neengxaki ezizodwa kwi-https://www.our-ai.org/ai-nexus/read! Le nqaku lithunyelwe kwinqanaba le-Our AI, i-AI Ethics organization eyenziwe ngama-students kunye ne-students eyenza ukwandisa iinkcukacha kwi-AI ngaphandle kwexesha elidlulileyo e-media. Ukuba ufumane le nqaku, nqakraza iimpazamo zethu yenyanga kunye neengxaki ezizodwa kwi-AI. ! https://www.our-ai.org/ai-nexus/read For a Greener Future Nangona i-validity ye-quantitative research eyenziwe ngexesha elidlulileyo ye-AI, ndingathanda ukuba kufuneka uqhagamshelane kwi-path to make better and better AI no matter what. Ukukhuthaza i-point that this must be done safely and with the intention of benefiting a reasonably large portion of society, ndingathanda ukuba i-avancement technology is often the only way to solve the world’s problems. Ngoko ke, i-engineering ye-genetic ingxakiwe i-seed for the Second Agricultural Revolution kulindeleke ekunciphiseni i-crisis yeengxaki ze-aliments e-South East Asia, kunye neevaccines that brought dozens of prevalent, pestilent diseases to their knees in a matter of years. Ngokutsho, nang

The is an opinion piece based on the author’s POV and does not necessarily reflect the views of HackerNoon.

Read My Stories

Le audio iveliswe ngolwimi lokuqala lwebali!

I-AI ayikho kakhulu ekhulwini njengoko ungathandayo

About Author

IZIMVO

ZIJONGE IIMPAWU

ELI NQAKU LINIKEZELWE KU

Related Stories

Light-Mode

Classic

Newspaper

Dark-Mode

Neon Noir

Minty

HN StartUps