汉字News

← 汉字NewsChinese News for Learners
technology
kē jì
科技
·HSK 5

Anthropic
Anthropic
Anthropic
restricts
xiànzhì
限制
release of
fābù
发布
new
xīnxíng
新型
cybersecurity
wǎngluò ānquán
网络安全
AI
AI
AI
model
móxíng
模型
。

Anthropic
Anthropic
Anthropic
released
fābù
发布
aspect marker
le
了
a
yī kuǎn
一款
named
míng wéi
名为
Claude Mythos Preview
Claude Mythos Preview
Claude Mythos Preview
possessive particle
de
的
new
xīn
新
AI
AI
AI
model
móxíng
模型
。
because
yóuyú
由于
its
qí
其
powerful
qiángdà
强大
possessive particle
de
的
vulnerability
lòudòng
漏洞
discovery
fāxiàn
发现
capability
nénglì
能力
，
the company
gōngsī
公司
only
zhǐ
只
to
xiàng
向
about
yuē
约
40 (companies)
sì shí jiā
40家
firms
qǐyè
企业
opened
kāifàng
开放
access
shǐyòng
使用
privileges
quánxiàn
权限
。

April 8, 2026 at 7:00 PM

artificial intelligence
réngōng zhìnéng
人工智能
company
gōngsī
公司
Anthropic
Anthropic
Anthropic
recently
jìnrì
近日
announced
xuānbù
宣布
the launch of
tuīchū
推出
a
yī kuǎn
一款
named
míng wéi
名为
Claude Mythos Preview
Claude Mythos Preview
Claude Mythos Preview
possessive particle
de
的
new-generation
xīnxíng
新型
AI
AI
AI
model
móxíng
模型
。
according to
jù
据
reports
bàodào
报道
，
this
gāi
该
model
móxíng
模型
in
zài
在
identifying
fāxiàn
发现
and
hé
和
exploiting
lìyòng
利用
software
ruǎnjiàn
软件
vulnerabilities
lòudòng
漏洞
in the area of
fāngmiàn
方面
possessive particle
de
的
capabilities
nénglì
能力
extremely
jíqí
极其
powerful
qiángdà
强大
。
therefore
yīncǐ
因此
，
the company
gōngsī
公司
decided
juédìng
决定
not
bù
不
publicly
gōngkāi
公开
release
fābù
发布
this
zhè kuǎn
这款
model
móxíng
模型
。
access
shǐyòng
使用
rights
quánxiàn
权限
will
jiāng
将
only
jǐn
仅
be opened
kāifàng
开放
to
gěi
给
about
yuē
约
40
sì shí jiā
40家
tech
kējì
科技
and
hé
和
cybersecurity
wǎngluò ānquán
网络安全
firms
qǐyè
企业
。
among them
qízhōng
其中
included
bāokuò
包括
Apple
Píngguǒ
苹果
、
Amazon
Yàmǎxùn
亚马逊
and
hé
和
Microsoft
Wēiruǎn
微软
among others
děng
等
large
dàxíng
大型
enterprises
qǐyè
企业
。
these
zhèxiē
这些
companies
gōngsī
公司
will
jiāng
将
use
lìyòng
利用
the
gāi
该
model
móxíng
模型
to
lái
来
detect
jiǎncè
检测
and
bìng
并
fix
xiūfù
修复
their own
zìshēn
自身
products
chǎnpǐn
产品
within
zhōng
中
possessive particle
de
的
security
ānquán
安全
risks
yǐnhuàn
隐患
。
Anthropic
Anthropic
Anthropic
stated
biǎoshì
表示
，
this move
cǐ jǔ
此举
aims to
zhǐ zài
旨在
prevent
bìmiǎn
避免
the technology
jìshù
技术
being
bèi
被
unlawful
bùfǎ
不法
actors
fènzǐ
分子
misused
lànyòng
滥用
。
industry
yènèi
业内
insiders
rénshì
人士
believe
rènwéi
认为
，
this
zhè
这
marks
biāozhì zhe
标志着
AI
AI
AI
in
zài
在
cybersecurity
wǎngluò ānquán
网络安全
domain
lǐngyù
领域
possessive particle
de
的
important
zhòngyào
重要
progress
jìnzhǎn
进展
。

New Words

réngōng zhìnéng
人工智能
artificial intelligence
móxíng
模型
model
lòudòng
漏洞
vulnerabilities
gōngkāi
公开
publicly
qǐyè
企业
enterprise
ānquán
安全
security

News in English

Science & Technology

Anthropic Restricts Access to Powerful New Cybersecurity AI Model

Anthropic has released a new AI model named Claude Mythos Preview. Due to its powerful vulnerability-discovery capabilities, the company has opened access to only about 40 enterprises.

Artificial intelligence company Anthropic recently announced the launch of a new-generation AI model named Claude Mythos Preview. According to reports, this model has extremely powerful capabilities in identifying and exploiting software vulnerabilities. Therefore, the company decided not to publicly release this model. Access rights will only be opened to about 40 technology and cybersecurity firms. These include large enterprises such as Apple, Amazon, and Microsoft. These companies will use the model to detect and fix security risks in their own products. Anthropic stated that this move aims to prevent the technology from being misused by unlawful actors. Industry insiders believe this marks important progress for AI in the cybersecurity domain.

More in Science & Technology

China
Zhongguó
中国
AI
réngong zhìnéng
人工智能
models
móxíng
模型
captures
zhànlǐng
占领
market
shìchǎng
市场

Chinese AI Models Gain Market Share as GLM 5.2 Rivals OpenAI and Anthropic at Low Cost

UBTech
Yōubìxuǎn
优必选
U1
U1
U1
humanoid robot
rénxíng jīqìrén
人形机器人
launches
fābuō
发布
price
jiàgé
价格

UBTech Launches U1 Ultra-Bionic Humanoid Robot Series Starting at 17600 Dollars

Apple (company)
Píngguǒ
苹果
to file a lawsuit / to sue
qǐsù
起诉
OpenAI (AI company)
OpenAI
OpenAI
，
to accuse / to charge
zhǐkòng
指控
it / the company (formal)
qí
其
to steal / to misappropriate
qièqǔ
窃取
hardware
yìngjiàn
硬件
trade secrets / confidential business information
shāngyè jīmì
商业机密
。

Apple Sues OpenAI Alleging Hardware Trade Secret Theft

Tencent (tech company)
Téngxùn
腾讯
in
zài
在
WeChat (messaging app)
Wēixìn
微信
to launch / roll out
tuīchū
推出
artificial intelligence
réngōng zhìnéng
人工智能
assistant
zhùshǒu
助手
Xiaowei (AI assistant name)
Xiǎowēi
小微
。

Tencent Begins Rolling Out AI Agent Xiaowei to WeChat Users