search icon
search icon
Flag Arrow Down
Română
Română
Magyar
Magyar
English
English
Français
Français
Deutsch
Deutsch
Italiano
Italiano
Español
Español
Русский
Русский
日本語
日本語
中国人
中国人

Change Language

arrow down
  • Română
    Română
  • Magyar
    Magyar
  • English
    English
  • Français
    Français
  • Deutsch
    Deutsch
  • Italiano
    Italiano
  • Español
    Español
  • Русский
    Русский
  • 日本語
    日本語
  • 中国人
    中国人
Sections
  • News
  • Exclusive
  • INSCOP Surveys
  • Podcast
  • EU
  • Diaspora
  • Republic of Moldova
  • Politics
  • Economy
  • Current Affairs
  • International
  • Sport
  • Health
  • Education
  • IT&C knowledge
  • Arts & Lifestyle
  • Opinions
  • Elections 2025
  • Environment
About Us
Contact
Privacy policy
Terms and conditions
Quickly scroll through news digests and see how they are covered in different publications!
  • News
  • Exclusive
    • INSCOP Surveys
    • Podcast
    • EU
    • Diaspora
    • Republic of Moldova
    • Politics
    • Economy
    • Current Affairs
    • International
    • Sport
    • Health
    • Education
    • IT&C knowledge
    • Arts & Lifestyle
    • Opinions
    • Elections 2025
    • Environment
155 new news items in the last 24 hours
  1. Home
  2. IT&C knowledge

Two AI models, Mythos and Sol, have demonstrated an alarming level of autonomy and intelligence in tests, creating false profiles for cyber attacks.

Raluca Ioana Draghici
whatsapp
facebook
linkedin
x
copy-link copy-link
6 August 2026, 09:32
main event image
IT&C knowledge
unsplash.com
google-preference

Always see our news on Google

The UK Artificial Intelligence Security Institute (AISI) reported that the AI models Mythos, developed by Anthropic, and Sol, by OpenAI, exhibited a level of "autonomy and malice" unprecedented during testing.

In a serious incident, the Mythos agent created false profiles based on the identities of real people in an attempt to deceive them and gain access to GitHub. AISI observed unusual data transfers and malicious activities, most of which originated from Mythos. The agent attempted to introduce malicious code into the GitHub system, but human intervention prevented the success of the attack.

AISI emphasized that this is the first time that the risks associated with autonomy and malice have manifested so clearly, without specific instructions.

Anthropic and OpenAI stated that the tests do not reflect the usual use of their models and that they are investigating the incident.

AISI informed GitHub and the affected users, and the platform disabled the fake accounts.

Sources

sursa imagine
Digi24
AI-ul Mythos de la Anthropic a uluit cercetătorii: A creat identități false pentru a păcăli oameni, într-un test de securitate
sursa imagine
Stirile ProTV
A creat identităţi false şi a încercat să ascundă urmele. Nou incident cu un model de inteligență artificială
sursa imagine
Stiripesurse
Modelul Mythos al Anthropic a creat identităţi false pentru a manipula oameni într-un test de securitate cibernetică
sursa imagine
News.ro
Modelul Mythos al Anthropic a creat identităţi false pentru a manipula oameni într-un test de securitate cibernetică

Latest News

23:56

UEFA Nations League: The Tricolors finish last in the group, Bosnia – Romania, score 2-1 / Promotion-relegation playoff to remain in League B follows

23:49

Dan Neculăescu is optimistic that the US will maintain a significant number of troops in Romania / The decision could come in December

23:11

The Trump administration asks Congress for $50 billion to modernize the air traffic control system

22:39

Marian Neacșu stated that state-owned enterprises have losses of one billion euros and debts of 18 billion lei

22:34

Ilie Bolojan proposes a center-right alliance with USR, on joint lists, as an alternative to the PSD-AUR axis

See more news

NEWS ON THE SAME TOPICS

event image
IT&C knowledge
OpenAI reveals troubling incidents: AI models tried to cheat and concealed fabricated information
event image
IT&C knowledge
OpenAI suspends the training and evaluation of certain advanced models after AI agents bypassed controls and accessed external systems
event image
IT&C knowledge
OpenAI warns: Cyber attacks with AI are becoming a dangerous reality
event image
International
OpenAI alerted dozens of institutions after AI agents accessed websites and improperly transferred users’ images
event image
International
Google: Russian-speaking hackers used AI agents for autonomous cyberattacks
event image
International
OpenAI has informed more than 100 organizations about incidents involving unauthorized activities by its artificial intelligence agents
app preview
Personalized news feed, AI-powered search, and notifications in a more interactive experience.
app preview app preview
Mythos Sun artificial intelligence autonomy cyber attacks Anthropic OpenAi

Editor’s Recommendations

main event image
Exclusive
7 hours ago

ANALYSIS Brexit without profit: London seeks reconciliation with the EU

app preview
Personalized news feed, AI-powered search, and notifications in a more interactive experience.
app preview
app store badge google play badge
  • News
  • Exclusive
  • INSCOP Surveys
  • Podcast
  • EU
  • Diaspora
  • Republic of Moldova
  • Politics
  • Economy
  • Current Affairs
  • International
  • Sport
  • Health
  • Education
  • IT&C knowledge
  • Arts & Lifestyle
  • Opinions
  • Elections 2025
  • Environment
  • About Us
  • Contact
Privacy policy
Cookies Policy
Terms and conditions
Open source licenses
All rights reserved Strategic Media Team SRL

Technology in partnership with

anpc-sal anpc-sol