Anthropic’s latest AI model detects far more software bugs than older tools. Microsoft, however, struggles to repair the issues quickly enough, raising concerns about software security and public safety.
Anthropic recently released an advanced AI system designed to scan code for hidden flaws. Early tests show it finds significantly more bugs than previous methods, including rare vulnerabilities that human reviewers often miss. The tool uses large language models trained on vast amounts of software data to spot patterns linked to security risks.
Microsoft maintains millions of lines of code across Windows, Office, and cloud services. Company engineers report that the new detections arrive faster than their teams can verify and patch them. This backlog leaves some systems exposed for weeks or months, increasing the chance that attackers will exploit the weaknesses before fixes are ready.
The gap between detection and repair affects everyday users who rely on Microsoft products for work and personal tasks. Experts warn that faster bug finding must be matched by faster fixes, or the public will face growing risks from data breaches and system failures. Both companies now face pressure to improve coordination and response times.
836.568 SRY