{"id":100,"date":"2023-11-15T15:24:54","date_gmt":"2023-11-15T08:24:54","guid":{"rendered":"https:\/\/xuhuongai.com\/?p=100"},"modified":"2023-11-18T12:18:16","modified_gmt":"2023-11-18T05:18:16","slug":"chatgpt-detector-co-the-bat-duoc-cac-bai-bao-duoc-tao-ra-boi-tri-tue-nhan-tao-voi-do-chinh-xac-chua-tung-co","status":"publish","type":"post","link":"https:\/\/xuhuongai.com\/?p=100","title":{"rendered":"“ChatGPT Detector” c\u00f3 th\u1ec3 b\u1eaft \u0111\u01b0\u1ee3c c\u00e1c b\u00e0i b\u00e1o t\u1ea1o ra b\u1edfi tr\u00ed tu\u1ec7 nh\u00e2n t\u1ea1o v\u1edbi \u0111\u1ed9 ch\u00ednh x\u00e1c ch\u01b0a t\u1eebng c\u00f3"},"content":{"rendered":"\n
\n

B\u1ea3n tin \u0111\u01b0\u1ee3c d\u1ecbch v\u00e0 t\u00f3m t\u1eaft b\u1edfi n\u1ec1n t\u1ea3ng t\u1ea1o tr\u1ee3 l\u00fd AI – KamiMind<\/a>.<\/p>\nNgu\u1ed3n: Mckenzie Prillaman & Nature Magazine, “\u2018ChatGPT Detector\u2019 Catches AI-Generated Papers with Unprecedented Accuracy<\/a>“, Scientific American, 07\/11\/2023.<\/cite><\/blockquote>\n\n\n

\n
\"\"
M\u1ed9t c\u00f4ng c\u1ee5 AI m\u1edbi c\u00f3 th\u1ec3 x\u00e1c \u0111\u1ecbnh ch\u00ednh x\u00e1c c\u00e1c b\u00e0i b\u00e1o chuy\u00ean ng\u00e0nh h\u00f3a h\u1ecdc \u0111\u01b0\u1ee3c vi\u1ebft b\u1edfi ChatGPT. Ngu\u1ed3n: Marco Bertorello\/AFP via Getty Images<\/figcaption><\/figure>\n<\/div>\n\n\n

M\u1ed9t c\u00f4ng c\u1ee5 m\u1edbi \u0111\u00e3 \u0111\u01b0\u1ee3c ph\u00e1t tri\u1ec3n s\u1eed d\u1ee5ng h\u1ecdc m\u00e1y \u0111\u1ec3 ph\u00e2n bi\u1ec7t gi\u1eefa vi\u1ec7c vi\u1ebft c\u1ee7a con ng\u01b0\u1eddi v\u00e0 tr\u00ed tu\u1ec7 nh\u00e2n t\u1ea1o d\u1ef1a tr\u00ean c\u00e1c \u0111\u1eb7c \u0111i\u1ec3m v\u1ec1 phong c\u00e1ch. C\u00f4ng c\u1ee5 n\u00e0y \u0111\u00e3 \u0111\u01b0\u1ee3c th\u1eed nghi\u1ec7m tr\u00ean c\u00e1c b\u00e0i vi\u1ebft v\u1ec1 h\u00f3a h\u1ecdc v\u00e0 cho th\u1ea5y kh\u1ea3 n\u0103ng nh\u1eadn di\u1ec7n cao h\u01a1n so v\u1edbi c\u00e1c c\u00f4ng c\u1ee5 hi\u1ec7n c\u00f3. C\u00f4ng c\u1ee5 n\u00e0y c\u00f3 th\u1ec3 gi\u00fap c\u00e1c nh\u00e0 xu\u1ea5t b\u1ea3n h\u1ecdc thu\u1eadt ph\u00e1t hi\u1ec7n c\u00e1c b\u00e0i vi\u1ebft \u0111\u01b0\u1ee3c t\u1ea1o ra b\u1edfi tr\u00ed tu\u1ec7 nh\u00e2n t\u1ea1o. Nghi\u00ean c\u1ee9u c\u0169ng \u0111\u00e3 ch\u1ec9 ra r\u1eb1ng vi\u1ec7c \u0111i\u1ec1u ch\u1ec9nh ph\u1ea7n m\u1ec1m cho c\u00e1c lo\u1ea1i v\u0103n b\u1ea3n c\u1ee5 th\u1ec3 c\u00f3 th\u1ec3 t\u0103ng c\u01b0\u1eddng kh\u1ea3 n\u0103ng ph\u00e1t hi\u1ec7n tr\u00ed tu\u1ec7 nh\u00e2n t\u1ea1o. Tuy nhi\u00ean, c\u1ea7n l\u01b0u \u00fd r\u1eb1ng vi\u1ec7c s\u1eed d\u1ee5ng c\u00f4ng c\u1ee5 n\u00e0y ch\u1ec9 t\u1eadp trung v\u00e0o m\u1ed9t l\u0129nh v\u1ef1c c\u1ee5 th\u1ec3 v\u00e0 kh\u00f4ng gi\u1ea3i quy\u1ebft \u0111\u01b0\u1ee3c c\u00e1c v\u1ea5n \u0111\u1ec1 r\u1ed9ng h\u01a1n li\u00ean quan \u0111\u1ebfn vi\u1ec7c vi\u1ebft b\u00e0i trong gi\u1edbi h\u1ecdc thu\u1eadt.<\/p>\n\n\n\n

B\u1ea3n t\u00f3m t\u1eaft ti\u1ebfng Anh<\/summary>\n

A new machine learning tool has been developed to differentiate between human and AI authors based on writing style. The tool was tested on chemistry papers written by the AI chatbot ChatGPT, and outperformed two existing AI detectors. This specialized classifier could assist academic publishers in identifying papers created by AI text generators. The study suggests that tailoring software to specific types of writing can enhance AI detection efforts. The tool analyzes 20 features of writing style, such as sentence length variation and word frequency, to determine authorship. When tested on chemistry journal introductions, the tool accurately identified sections written by ChatGPT with 100% accuracy based on titles and 98% accuracy based on abstracts. In contrast, existing AI detectors performed significantly worse. While the tool is specialized for scientific journal articles, it struggled to recognize articles from university newspapers as human-written. The use of writing style features for authorship detection is a novel approach and complements existing methods that rely on predictive text patterns. However, it is important to recognize that AI detection tools alone cannot address broader issues in academia, such as the pressure to produce papers quickly.<\/p>\n<\/details>\n\n\n\n

B\u1ea3n d\u1ecbch Anh – Vi\u1ec7t<\/summary>\n

M\u1ed9t c\u00f4ng c\u1ee5 m\u1edbi d\u1ef1a tr\u00ean h\u1ecdc m\u00e1y s\u1eed d\u1ee5ng c\u00e1c \u0111\u1eb7c \u0111i\u1ec3m v\u1ec1 phong c\u00e1ch vi\u1ebft \u0111\u1ec3 ph\u00e2n bi\u1ec7t gi\u1eefa c\u00e1c t\u00e1c gi\u1ea3 con ng\u01b0\u1eddi v\u00e0 tr\u00ed tu\u1ec7 nh\u00e2n t\u1ea1o.
Theo m\u1ed9t nghi\u00ean c\u1ee9u \u0111\u01b0\u1ee3c c\u00f4ng b\u1ed1 v\u00e0o ng\u00e0y 6 th\u00e1ng 11 tr\u00ean t\u1ea1p ch\u00ed Cell Reports Physical Science, m\u1ed9t c\u00f4ng c\u1ee5 h\u1ecdc m\u00e1y c\u00f3 th\u1ec3 d\u1ec5 d\u00e0ng nh\u1eadn ra khi c\u00e1c b\u00e0i vi\u1ebft v\u1ec1 h\u00f3a h\u1ecdc \u0111\u01b0\u1ee3c vi\u1ebft b\u1eb1ng chatbot ChatGPT. B\u1ed9 ph\u00e2n lo\u1ea1i chuy\u00ean bi\u1ec7t n\u00e0y \u0111\u00e3 v\u01b0\u1ee3t tr\u1ed9i h\u01a1n hai b\u1ed9 ph\u00e1t hi\u1ec7n tr\u00ed tu\u1ec7 nh\u00e2n t\u1ea1o hi\u1ec7n c\u00f3 v\u00e0 c\u00f3 th\u1ec3 gi\u00fap c\u00e1c nh\u00e0 xu\u1ea5t b\u1ea3n h\u1ecdc thu\u1eadt x\u00e1c \u0111\u1ecbnh \u0111\u01b0\u1ee3c c\u00e1c b\u00e0i vi\u1ebft \u0111\u01b0\u1ee3c t\u1ea1o ra b\u1edfi c\u00e1c tr\u00ecnh t\u1ea1o v\u0103n b\u1ea3n tr\u00ed tu\u1ec7 nh\u00e2n t\u1ea1o.<\/p>\n\n\n\n

“Heather Desaire, m\u1ed9t nh\u00e0 h\u00f3a h\u1ecdc t\u1ea1i \u0110\u1ea1i h\u1ecdc Kansas \u1edf Lawrence, ng\u01b0\u1eddi \u0111\u1ed3ng t\u00e1c gi\u1ea3 cho bi\u1ebft: “H\u1ea7u h\u1ebft c\u00e1c l\u0129nh v\u1ef1c ph\u00e2n t\u00edch v\u0103n b\u1ea3n mu\u1ed1n c\u00f3 m\u1ed9t b\u1ed9 ph\u00e1t hi\u1ec7n t\u1ed5ng qu\u00e1t th\u1ef1c s\u1ef1 c\u00f3 th\u1ec3 ho\u1ea1t \u0111\u1ed9ng tr\u00ean m\u1ecdi th\u1ee9”. Nh\u01b0ng b\u1eb1ng c\u00e1ch t\u1ea1o ra m\u1ed9t c\u00f4ng c\u1ee5 t\u1eadp trung v\u00e0o m\u1ed9t lo\u1ea1i b\u00e0i vi\u1ebft c\u1ee5 th\u1ec3, “ch\u00fang t\u00f4i th\u1ef1c s\u1ef1 \u0111ang nh\u1eafm t\u1edbi s\u1ef1 ch\u00ednh x\u00e1c”.
Nh\u1eefng k\u1ebft qu\u1ea3 n\u00e0y cho th\u1ea5y vi\u1ec7c ph\u00e1t tri\u1ec3n b\u1ed9 ph\u00e1t hi\u1ec7n tr\u00ed tu\u1ec7 nh\u00e2n t\u1ea1o c\u00f3 th\u1ec3 \u0111\u01b0\u1ee3c t\u0103ng c\u01b0\u1eddng b\u1eb1ng c\u00e1ch \u0111i\u1ec1u ch\u1ec9nh ph\u1ea7n m\u1ec1m cho c\u00e1c lo\u1ea1i v\u0103n b\u1ea3n c\u1ee5 th\u1ec3, Desaire n\u00f3i. “N\u1ebfu b\u1ea1n c\u00f3 th\u1ec3 x\u00e2y d\u1ef1ng m\u1ed9t c\u00e1i g\u00ec \u0111\u00f3 nhanh ch\u00f3ng v\u00e0 d\u1ec5 d\u00e0ng, th\u00ec kh\u00f4ng kh\u00f3 \u0111\u1ec3 x\u00e2y d\u1ef1ng m\u1ed9t c\u00e1i g\u00ec \u0111\u00f3 cho c\u00e1c l\u0129nh v\u1ef1c kh\u00e1c”.<\/p>\n\n\n\n

C\u00c1C Y\u1ebeU T\u1ed0 V\u1ec0 PHONG C\u00c1CH<\/p>\n\n\n\n

Desaire v\u00e0 \u0111\u1ed3ng nghi\u1ec7p c\u1ee7a c\u00f4 \u0111\u00e3 m\u00f4 t\u1ea3 c\u00f4ng c\u1ee5 ph\u00e1t hi\u1ec7n ChatGPT c\u1ee7a h\u1ecd v\u00e0o th\u00e1ng 6, khi h\u1ecd \u00e1p d\u1ee5ng n\u00f3 cho c\u00e1c b\u00e0i vi\u1ebft quan \u0111i\u1ec3m t\u1eeb t\u1ea1p ch\u00ed Science. S\u1eed d\u1ee5ng h\u1ecdc m\u00e1y, c\u00f4ng c\u1ee5 ph\u00e1t hi\u1ec7n xem x\u00e9t 20 \u0111\u1eb7c \u0111i\u1ec3m v\u1ec1 phong c\u00e1ch vi\u1ebft, bao g\u1ed3m s\u1ef1 bi\u1ebfn \u0111\u1ed5i trong \u0111\u1ed9 d\u00e0i c\u00e2u, v\u00e0 t\u1ea7n su\u1ea5t c\u1ee7a m\u1ed9t s\u1ed1 t\u1eeb v\u00e0 d\u1ea5u ch\u1ea5m c\u00e2u c\u1ee5 th\u1ec3, \u0111\u1ec3 x\u00e1c \u0111\u1ecbnh xem m\u1ed9t nh\u00e0 khoa h\u1ecdc h\u1ecdc thu\u1eadt hay ChatGPT \u0111\u00e3 vi\u1ebft m\u1ed9t \u0111o\u1ea1n v\u0103n. Nh\u1eefng k\u1ebft qu\u1ea3 cho th\u1ea5y “b\u1ea1n c\u00f3 th\u1ec3 s\u1eed d\u1ee5ng m\u1ed9t t\u1eadp h\u1ee3p nh\u1ecf c\u00e1c \u0111\u1eb7c \u0111i\u1ec3m \u0111\u1ec3 \u0111\u1ea1t \u0111\u01b0\u1ee3c m\u1ee9c \u0111\u1ed9 ch\u00ednh x\u00e1c cao”, Desaire n\u00f3i.<\/p>\n\n\n\n

Trong nghi\u00ean c\u1ee9u m\u1edbi nh\u1ea5t, c\u00f4ng c\u1ee5 \u0111\u00e3 \u0111\u01b0\u1ee3c hu\u1ea5n luy\u1ec7n tr\u00ean c\u00e1c ph\u1ea7n gi\u1edbi thi\u1ec7u c\u1ee7a c\u00e1c b\u00e0i vi\u1ebft t\u1eeb m\u01b0\u1eddi t\u1ea1p ch\u00ed h\u00f3a h\u1ecdc do Hi\u1ec7p h\u1ed9i H\u00f3a h\u1ecdc M\u1ef9 (ACS) xu\u1ea5t b\u1ea3n. Nh\u00f3m nghi\u00ean c\u1ee9u \u0111\u00e3 ch\u1ecdn ph\u1ea7n gi\u1edbi thi\u1ec7u v\u00ec ph\u1ea7n n\u00e0y c\u1ee7a m\u1ed9t b\u00e0i vi\u1ebft kh\u00e1 d\u1ec5 cho ChatGPT vi\u1ebft n\u1ebfu n\u00f3 c\u00f3 truy c\u1eadp v\u00e0o t\u00e0i li\u1ec7u n\u1ec1n, Desaire n\u00f3i. C\u00e1c nh\u00e0 nghi\u00ean c\u1ee9u \u0111\u00e3 hu\u1ea5n luy\u1ec7n c\u00f4ng c\u1ee5 c\u1ee7a h\u1ecd tr\u00ean 100 ph\u1ea7n gi\u1edbi thi\u1ec7u \u0111\u01b0\u1ee3c xu\u1ea5t b\u1ea3n \u0111\u1ec3 ph\u1ee5c v\u1ee5 l\u00e0m v\u0103n b\u1ea3n do con ng\u01b0\u1eddi vi\u1ebft, sau \u0111\u00f3 y\u00eau c\u1ea7u ChatGPT-3.5 vi\u1ebft 200 ph\u1ea7n gi\u1edbi thi\u1ec7u theo ki\u1ec3u t\u1ea1p ch\u00ed ACS. \u0110\u1ed1i v\u1edbi 100 ph\u1ea7n n\u00e0y, c\u00f4ng c\u1ee5 \u0111\u01b0\u1ee3c cung c\u1ea5p v\u1edbi ti\u00eau \u0111\u1ec1 c\u1ee7a b\u00e0i b\u00e1o, v\u00e0 \u0111\u1ed1i v\u1edbi 100 ph\u1ea7n c\u00f2n l\u1ea1i, n\u00f3 \u0111\u01b0\u1ee3c cung c\u1ea5p v\u1edbi t\u00f3m t\u1eaft c\u1ee7a ch\u00fang.<\/p>\n\n\n\n

Khi \u0111\u01b0\u1ee3c th\u1eed nghi\u1ec7m tr\u00ean c\u00e1c ph\u1ea7n gi\u1edbi thi\u1ec7u \u0111\u01b0\u1ee3c vi\u1ebft b\u1edfi con ng\u01b0\u1eddi v\u00e0 nh\u1eefng ph\u1ea7n \u0111\u01b0\u1ee3c t\u1ea1o ra b\u1edfi tr\u00ed tu\u1ec7 nh\u00e2n t\u1ea1o t\u1eeb c\u00e1c t\u1ea1p ch\u00ed c\u00f9ng lo\u1ea1i, c\u00f4ng c\u1ee5 \u0111\u00e3 nh\u1eadn d\u1ea1ng c\u00e1c ph\u1ea7n \u0111\u01b0\u1ee3c vi\u1ebft b\u1edfi ChatGPT-3.5 d\u1ef1a tr\u00ean ti\u00eau \u0111\u1ec1 v\u1edbi \u0111\u1ed9 ch\u00ednh x\u00e1c 100%. \u0110\u1ed1i v\u1edbi c\u00e1c ph\u1ea7n gi\u1edbi thi\u1ec7u \u0111\u01b0\u1ee3c t\u1ea1o ra b\u1edfi ChatGPT d\u1ef1a tr\u00ean t\u00f3m t\u1eaft, \u0111\u1ed9 ch\u00ednh x\u00e1c h\u01a1i th\u1ea5p h\u01a1n, \u1edf m\u1ee9c 98%. C\u00f4ng c\u1ee5 ho\u1ea1t \u0111\u1ed9ng t\u1ed1t v\u1edbi v\u0103n b\u1ea3n \u0111\u01b0\u1ee3c vi\u1ebft b\u1edfi ChatGPT-4, phi\u00ean b\u1ea3n m\u1edbi nh\u1ea5t c\u1ee7a chatbot. Tr\u00e1i l\u1ea1i, b\u1ed9 ph\u00e1t hi\u1ec7n tr\u00ed tu\u1ec7 nh\u00e2n t\u1ea1o ZeroGPT ch\u1ec9 nh\u1eadn d\u1ea1ng c\u00e1c ph\u1ea7n gi\u1edbi thi\u1ec7u \u0111\u01b0\u1ee3c vi\u1ebft b\u1edfi tr\u00ed tu\u1ec7 nh\u00e2n t\u1ea1o v\u1edbi \u0111\u1ed9 ch\u00ednh x\u00e1c ch\u1ec9 kho\u1ea3ng 35-65%, ph\u1ee5 thu\u1ed9c v\u00e0o phi\u00ean b\u1ea3n c\u1ee7a ChatGPT \u0111\u01b0\u1ee3c s\u1eed d\u1ee5ng v\u00e0 xem li\u1ec7u ph\u1ea7n gi\u1edbi thi\u1ec7u \u0111\u01b0\u1ee3c t\u1ea1o ra t\u1eeb ti\u00eau \u0111\u1ec1 hay t\u00f3m t\u1eaft c\u1ee7a b\u00e0i b\u00e1o. M\u1ed9t c\u00f4ng c\u1ee5 ph\u00e2n lo\u1ea1i v\u0103n b\u1ea3n \u0111\u01b0\u1ee3c s\u1ea3n xu\u1ea5t b\u1edfi OpenAI, nh\u00e0 s\u1ea3n xu\u1ea5t ChatGPT, c\u0169ng ho\u1ea1t \u0111\u1ed9ng k\u00e9m – n\u00f3 c\u00f3 th\u1ec3 nh\u1eadn ra c\u00e1c ph\u1ea7n gi\u1edbi thi\u1ec7u \u0111\u01b0\u1ee3c vi\u1ebft b\u1edfi tr\u00ed tu\u1ec7 nh\u00e2n t\u1ea1o v\u1edbi \u0111\u1ed9 ch\u00ednh x\u00e1c kho\u1ea3ng 10-55%.
C\u00f4ng c\u1ee5 ChatGPT m\u1edbi ngay c\u1ea3 ho\u1ea1t \u0111\u1ed9ng t\u1ed1t v\u1edbi c\u00e1c ph\u1ea7n gi\u1edbi thi\u1ec7u t\u1eeb c\u00e1c t\u1ea1p ch\u00ed n\u00f3 kh\u00f4ng \u0111\u01b0\u1ee3c hu\u1ea5n luy\u1ec7n, v\u00e0 n\u00f3 nh\u1eadn d\u1ea1ng v\u0103n b\u1ea3n tr\u00ed tu\u1ec7 nh\u00e2n t\u1ea1o \u0111\u01b0\u1ee3c t\u1ea1o ra t\u1eeb nhi\u1ec1u \u0111\u1ec1 m\u1ee5c kh\u00e1c nhau, bao g\u1ed3m c\u1ea3 m\u1ed9t \u0111\u1ec1 m\u1ee5c nh\u1eb1m l\u00e0m r\u1ed1i c\u00e1c b\u1ed9 ph\u00e1t hi\u1ec7n tr\u00ed tu\u1ec7 nh\u00e2n t\u1ea1o. Tuy nhi\u00ean, h\u1ec7 th\u1ed1ng n\u00e0y chuy\u00ean bi\u1ec7t cao cho b\u00e0i vi\u1ebft khoa h\u1ecdc trong c\u00e1c t\u1ea1p ch\u00ed. Khi \u0111\u01b0\u1ee3c tr\u00ecnh b\u00e0y v\u1edbi c\u00e1c b\u00e0i vi\u1ebft th\u1ef1c t\u1ebf t\u1eeb b\u00e1o \u0111\u1ea1i h\u1ecdc, n\u00f3 kh\u00f4ng nh\u1eadn ra ch\u00fang l\u00e0 do con ng\u01b0\u1eddi vi\u1ebft.<\/p>\n\n\n\n

C\u00c1C V\u1ea4N \u0110\u1ec0 R\u1ed8NG H\u01a0N<\/p>\n\n\n\n

Debora Weber-Wulff, m\u1ed9t nh\u00e0 khoa h\u1ecdc m\u00e1y t\u00ednh nghi\u00ean c\u1ee9u v\u1ec1 vi\u1ec7c sao ch\u00e9p h\u1ecdc thu\u1eadt t\u1ea1i \u0110\u1ea1i h\u1ecdc C\u00f4ng ngh\u1ec7 Berlin HTW, cho bi\u1ebft nh\u1eefng g\u00ec t\u00e1c gi\u1ea3 \u0111ang l\u00e0m l\u00e0 “m\u1ed9t \u0111i\u1ec1u th\u00fa v\u1ecb”. Theo c\u00f4, nhi\u1ec1u c\u00f4ng c\u1ee5 hi\u1ec7n c\u00f3 c\u1ed1 g\u1eafng x\u00e1c \u0111\u1ecbnh t\u00e1c gi\u1ea3 b\u1eb1ng c\u00e1ch t\u00ecm ki\u1ebfm c\u00e1c m\u1eabu v\u0103n b\u1ea3n c\u00f3 t\u00ednh d\u1ef1 \u0111o\u00e1n c\u1ee7a vi\u1ec7c vi\u1ebft tr\u00ed tu\u1ec7 nh\u00e2n t\u1ea1o thay v\u00ec xem x\u00e9t c\u00e1c \u0111\u1eb7c \u0111i\u1ec3m v\u1ec1 phong c\u00e1ch vi\u1ebft. “T\u00f4i ch\u01b0a bao gi\u1edd ngh\u0129 \u0111\u1ebfn vi\u1ec7c s\u1eed d\u1ee5ng stylometrics tr\u00ean ChatGPT”.<\/p>\n\n\n\n

Tuy nhi\u00ean, Weber-Wulff ch\u1ec9 ra r\u1eb1ng c\u00f3 c\u00e1c v\u1ea5n \u0111\u1ec1 kh\u00e1c \u0111ang th\u00fac \u0111\u1ea9y vi\u1ec7c s\u1eed d\u1ee5ng ChatGPT trong gi\u1edbi h\u1ecdc thu\u1eadt. C\u00f4 l\u01b0u \u00fd r\u1eb1ng nhi\u1ec1u nh\u00e0 nghi\u00ean c\u1ee9u \u0111ang ph\u1ea3i \u0111\u1ed1i m\u1eb7t v\u1edbi \u00e1p l\u1ef1c \u0111\u1ec3 nhanh ch\u00f3ng \u0111\u01b0a ra c\u00e1c b\u00e0i vi\u1ebft, ho\u1eb7c h\u1ecd c\u00f3 th\u1ec3 kh\u00f4ng coi vi\u1ec7c vi\u1ebft b\u00e0i l\u00e0 m\u1ed9t ph\u1ea7n quan tr\u1ecdng c\u1ee7a khoa h\u1ecdc. C\u00e1c c\u00f4ng c\u1ee5 ph\u00e1t hi\u1ec7n tr\u00ed tu\u1ec7 nh\u00e2n t\u1ea1o s\u1ebd kh\u00f4ng gi\u1ea3i quy\u1ebft \u0111\u01b0\u1ee3c nh\u1eefng v\u1ea5n \u0111\u1ec1 n\u00e0y, v\u00e0 kh\u00f4ng n\u00ean \u0111\u01b0\u1ee3c coi l\u00e0 “m\u1ed9t gi\u1ea3i ph\u00e1p ph\u1ea7n m\u1ec1m k\u1ef3 di\u1ec7u cho m\u1ed9t v\u1ea5n \u0111\u1ec1 x\u00e3 h\u1ed9i”.<\/p>\n<\/details>\n","protected":false},"excerpt":{"rendered":"

M\u1ed9t c\u00f4ng c\u1ee5 m\u1edbi \u0111\u00e3 \u0111\u01b0\u1ee3c ph\u00e1t tri\u1ec3n s\u1eed d\u1ee5ng h\u1ecdc m\u00e1y \u0111\u1ec3 ph\u00e2n bi\u1ec7t gi\u1eefa vi\u1ec7c vi\u1ebft c\u1ee7a con ng\u01b0\u1eddi v\u00e0 tr\u00ed tu\u1ec7 nh\u00e2n t\u1ea1o d\u1ef1a tr\u00ean c\u00e1c \u0111\u1eb7c \u0111i\u1ec3m v\u1ec1 phong c\u00e1ch.<\/p>\n","protected":false},"author":2,"featured_media":0,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[5],"tags":[17,10],"class_list":["post-100","post","type-post","status-publish","format-standard","hentry","category-ai-news","tag-hoc-may","tag-tri-tue-nhan-tao"],"_links":{"self":[{"href":"https:\/\/xuhuongai.com\/index.php?rest_route=\/wp\/v2\/posts\/100","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/xuhuongai.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/xuhuongai.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/xuhuongai.com\/index.php?rest_route=\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/xuhuongai.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=100"}],"version-history":[{"count":7,"href":"https:\/\/xuhuongai.com\/index.php?rest_route=\/wp\/v2\/posts\/100\/revisions"}],"predecessor-version":[{"id":442,"href":"https:\/\/xuhuongai.com\/index.php?rest_route=\/wp\/v2\/posts\/100\/revisions\/442"}],"wp:attachment":[{"href":"https:\/\/xuhuongai.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=100"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/xuhuongai.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=100"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/xuhuongai.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=100"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}