Deepseak is a 'good apricot'? Nobel Prize winner: "Exaggerated craze"

Demis Hassabis, CEO of Google DeepMind and winner of the Nobel Prize in Chemistry last year, said about the artificial intelligence (AI) of Chinese startup DeepSec, "There is no scientific progress and the hype is overblown." Demis Hassabis, CEO



of Google DeepMind, attending the 'AI Action

Summit' in Paris [Photo = AP/Yonhap News] According to a CNBC report on the 9th (local time), CEO Hassabis said about DeepSec at an event hosted by Google ahead of the AI ​​International Summit held in Paris, France that "It shows that you can do very good engineering and that you can change things on a geopolitical scale."



However, from a technical perspective, it is not a big change.



CEO Hassabis said, "There is no real new scientific progress, they are just using known technologies in the field of AI," and evaluated that the hype surrounding DeepSec is "a little overblown."



Regarding the timing of the emergence of human-level or higher AGI, he said, "I think it's close," and that it may be about five years away. He



also said, "Society needs to prepare for it," and "We need to make sure that we benefit from it and that society as a whole benefits, while also mitigating some of the risks."





https://www.inews24.com/view/blogger/1812077

댓글

댓글 쓰기

이 블로그의 인기 게시물

"TSMC 3nm Process Reaches 100% Utilization Rate in Shortest Time Ever"

이미지
Taiwan Semiconductor Foundry Manufacturing Company (TSMC) has recorded its fastest-ever full-scale operation rate of its advanced process. Taiwan TSMC logo [Photo = Yonhap News] According to Counterpoint Research on the 21st, TSMC's 3nm process reached a 100% operation rate for the first time in five quarters since mass production. This is due to a surge in demand for x86 PC CPUs and other application processors (AP SoCs), including Apple's A17 Pro and A18 Pro. Counterpoint Research forecasts that this high operation rate will continue as demand for artificial intelligence (AI) and high-performance computing (HPC) increases with the full-scale introduction of AI ASICs, such as Nvidia's Rubin graphics processing unit (GPU), Google's TPU v7, and AWS's Trainium 3. Older processes, such as 6/7nm and 4/5nm, were mainly focused on the smartphone market, so production expansion was relatively slow. The 6·7nm process utilization r...
자세한 내용 보기

K-Robot 'Ellis' Invited to Computex by NVIDIA

이미지
Korean:The Korean humanoid robot 'Ellis' met visitors at the Taiwan ICT exhibition 'Computex 2025'. According to the robot industry on the 21st, A-Robot is showcasing Ellis at the Nvidia exhibition hall in Computex, which is being held in Taipei, Taiwan until the 23rd. A-Robot's humanoid robot 'Ellis' hands out water to visitors after playing dice with them at Computex 2025 on the 21st. [Photo = A-Robot] A-Robot CEO Eom Yoon-seol (top left) introduces 'Ellis' to visitors at 'Computex 2025' held at the Taipei Nangang Exhibition Center in Taiwan on the 21st. Visitors are delighted to see Ellis' dice numbers revealed. [Photo = A-Robot] Ellis is a humanoid robot that is 160 cm tall and weighs 45 kg. It has been learning various human behaviors using Nvidia's Robot Foundation Model Cosmos platform. At this exhibition, A-Robot provided an opportunity for Ellis to interact with visitors. Ellis and ...
자세한 내용 보기

"TSMC 3nm Process Reaches 100% Utilization Rate in Shortest Time Ever"

이미지
Taiwan Semiconductor Foundry Manufacturing Company (TSMC) has recorded its fastest-ever full-scale operation rate of its advanced process. Taiwan TSMC logo [Photo = Yonhap News] According to Counterpoint Research on the 21st, TSMC's 3nm process reached a 100% operation rate for the first time in five quarters since mass production. This is due to a surge in demand for x86 PC CPUs and other application processors (AP SoCs), including Apple's A17 Pro and A18 Pro. Counterpoint Research forecasts that this high operation rate will continue as demand for artificial intelligence (AI) and high-performance computing (HPC) increases with the full-scale introduction of AI ASICs, such as Nvidia's Rubin graphics processing unit (GPU), Google's TPU v7, and AWS's Trainium 3. Older processes, such as 6/7nm and 4/5nm, were mainly focused on the smartphone market, so production expansion was relatively slow. The 6·7nm process utilization r...
자세한 내용 보기