🕸️BPSC WEB🕸️

#NCERT_CONQUEST #HISTORY

Anthropic’s new AI model “Mythos” • The model is part of the Claude AI family • It is a “vulnerability discovery AI”: • Can identify software bugs • But also exploit them (cyber attacks) Example: • Found flaw in Firefox JavaScript engine • Could gain system-level control (shell access) ➡️ Core theme: AI shifting from passive tool —> active cyber capability (offence + defence) ➡️ Mythos AI shows that modern AI is no longer just a tool to detect cyber problems - it can also automatically exploit them like a hacker. ➡️ This means: • AI can act like both a security expert AND a cyber attacker • Cyber attacks can become faster, automated, and large-scale • Even non-experts can carry out complex hacking using AI ➡️ The real problem: • Technology is advancing faster than laws and regulations Dual-use” Nature of AI • Same AI can: • ✅ Fix vulnerabilities • ❌ Exploit vulnerabilities