Tiga kerentanan AI berisiko tinggi yang ditemukan di Claude.ai

Indonesia

Para peneliti telah mengidentifikasi tiga kerentanan berisiko tinggi di Claude.ai. Hal ini memungkinkan rantai serangan ujung ke ujung yang mengeksfiltrasi informasi sensitif tanpa sepengetahuan pengguna. Iklan Google yang sah dapat memicu eksfiltrasi data.

TechRadar melaporkan pada 19 Maret 2026, penemuan tiga kerentanan AI berisiko tinggi di Claude.ai. Cacat tersebut membentuk rantai serangan ujung ke ujung yang mampu menyusupkan informasi sensitif tanpa sepengetahuan pengguna. Khususnya, iklan Google yang sah dapat menyebabkan eksfiltrasi data tersebut. Masalah keamanan ini menyoroti risiko dalam sistem AI di mana elemen eksternal seperti iklan dapat membahayakan data pengguna.

Artikel Terkait

Illustration of Anthropic restricting Claude Mythos AI and launching Project Glasswing consortium with tech giants to address cybersecurity vulnerabilities.
Gambar dihasilkan oleh AI

Anthropic restricts Claude Mythos AI release and launches Project Glasswing over cybersecurity risks

Dilaporkan oleh AI Gambar dihasilkan oleh AI

Anthropic has limited access to its Claude Mythos Preview AI model due to its superior ability to detect and exploit software vulnerabilities, while launching Project Glasswing—a consortium with over 45 tech firms including Apple, Google, and Microsoft—to collaboratively patch flaws and bolster defenses. The announcement follows recent data leaks at the firm.

Hackers create fake Claude site to spread malware

Cybersecurity researchers have identified a fraudulent website mimicking the popular AI tool Claude that delivers backdoor malware to visitors. The discovery highlights how cybercriminals are capitalizing on growing interest in artificial intelligence platforms.

Claude Mythos leak heightens cyber threats for banks

Dilaporkan oleh AI

Anthropic's latest AI model Claude Mythos has leaked despite being deemed too dangerous for public release. Financial institutions now face advanced AI-powered attacks capable of exploiting unknown vulnerabilities.

Technology

Security researchers breach macOS using Anthropic AI tool

Europe

BaFin echoes US warnings on Claude Mythos AI risks to banks

Technology

AI decodes Facebook ads into personal profiles

Anthropic's Mythos AI model sparks hacking fears

Anthropic has released a new cyber-focused AI model called Mythos, capable of detecting software flaws faster than humans and generating exploits. The model has raised alarms among governments and companies for potentially turbocharging hacking by exposing vulnerabilities quicker than they can be patched. Officials worldwide are scrambling to assess the risks.

AI trainers use chatbots to complete model tasks

Dilaporkan oleh AI

Workers paid to train advanced AI models are increasingly relying on chatbots like ChatGPT to generate the required conversations and tests. This shortcut, described as widespread by multiple sources, risks degrading the quality of future models through recursive training on synthetic data.

22 Mei 2026 04.20

Nearly all firms admit shipping vulnerable AI-generated code

12 Mei 2026 07.58

Scammers expand Claude malware campaign to target Mac users via ads and fake support sites

10 April 2026 01.15

US Treasury warns banks of AI cyberattack risks following Anthropic's Claude Mythos announcement

05 April 2026 04.58

Anthropic ends unlimited Claude access via third-party agents, requires extra payments for heavy use

31 Maret 2026 02.54

UK study reveals AI agents evading safeguards in user interactions

Situs web ini menggunakan cookie

Kami menggunakan cookie untuk analisis guna meningkatkan situs kami. Baca kebijakan privasi kami untuk informasi lebih lanjut.
Tolak