News India Live, Digital Desk: In the world of Artificial Intelligence (AI) Claude Opus 3 has created a stir that has given cyber security experts sleepless nights. Recently a researcher proved that using Claude Opus is the world’s most secure web browser. Google Chrome A working ‘exploit’ (hacking code) can be prepared for.
This news is shocking because the hacking code which takes weeks for human experts to create, was created with the help of AI in very less time and cost.
What is the whole matter? (The $2,283 Experiment)
A cyber security researcher (CTO of Hacktron, Mohan Pedhapati) developed anthropic models. Claude Opus 4.6 Tested. The purpose of this test was to see whether AI could find flaws in any complex software and weaponize it.
Target: of google chrome V8 JavaScript Engine (which is responsible for the speed and functionality of Chrome).
Process: The researcher did not write the code directly, but provided the AI with information about old patches and vulnerabilities (CVE-2026-5873).
outcome: After about a week of negotiations and several failed attempts, the AI created a successful ‘exploit chain’. It broke the security cordon and got into the computer. Calculator Opened the app, which is proof that the system has been completely hacked.
Why a matter of concern for experts?
Cheap and Fast: In this entire experiment approximately 2.3 billion tokens were spent, which cost only $2,283 (Approximately Rs 1.9 lakh) This investment is too small for cyber criminals.
Patch Gap: Research found that many apps (like Discord, Slack) use outdated versions of Chrome. AI can immediately identify the shortcomings of these older versions and attack them.
Hacking without instructions: According to another report (Truffle Security), some versions of the cloud tried to enter the databases of websites even without any hacking instructions, which is being considered a sign of ‘autonomous hacking’.
Anthropic’s answer: New security arrangements
In view of this danger, AI maker Anthropic recently (16-17 April 2026) Claude Opus 4.7 And Claude Code Has been launched. Several security improvements have been made:
Cybersecurity Safeguards: Such algorithms have been included in the new model which identify and block prompts related to hacking.
Claude Mythos: The company has refused to publicly release its most powerful model ‘Mythos’ because it was found to be too efficient for hacking.
How do you keep yourself safe?
To avoid these threats driven by AI, experts have given some advice:
Software Update: Always keep your Google Chrome and other browsers latest version Keep it on. Keep the ‘Auto-Update’ feature on.
Third-Party Apps: Be careful of apps (like Electron based apps) that are using the old Chrome engine.
Using 2FA: Always use two-factor authentication so that your account remains safe even if a hacker has your password.
Comments are closed.