⚡️ NEW: Anthropic introduced Claude Code Security, a web tool that scans code for vulnerabilities and suggests fixes, now in limited research preview. https://t.co/rtqQ0gvSW7 ...

RT Claude (@claudeai)Introducing Claude Code Security, now in limited research preview.It scans codebases for vulnerabilities and suggests targeted software patches for human review, allowing teams to find and fix issues that traditional tools often miss.Learn more: https://t.co/n4SZ9EIklG https://t.co/zw9NjpqFz9 ...

Core Viewpoint - Major tech companies are increasingly restricting the use of third-party AI programming tools to enhance code and data security, even at the cost of development efficiency [1][4][13]. Group 1: Company Actions - Kuaishou has tightened the usage rights of several third-party programming tools, leading to significant disruptions in development efficiency for engineers who relied on AI tools like Cursor [1]. - ByteDance was one of the first companies to implement systematic measures, announcing a ban on third-party AI programming software to prevent potential data leakage risks [5]. - Microsoft has also prohibited the use of unapproved AI services, emphasizing the importance of protecting intellectual property and customer trust [5]. - Amazon has directed engineers to prioritize the use of its in-house AI coding tool, Kiro, while discontinuing support for new third-party AI development tools [6]. Group 2: Industry Trends - The trend of restricting third-party AI tools reflects a broader industry shift towards prioritizing data sovereignty and security over short-term efficiency gains [4][13]. - Many tech giants are adopting internal AI tools to maintain control over their code, marking a significant change in the development landscape where "code is an asset" [4][6]. - The historical context shows that concerns over code and data security have existed long before the AI era, with companies implementing protective measures against potential data leaks [2][3]. Group 3: Employee Perspectives - Employees express frustration over the inefficiency of internal AI tools compared to their external counterparts, highlighting a significant drop in productivity [7][8]. - There is a growing sentiment among engineers that the pursuit of absolute security may hinder innovation and lead to missed opportunities for productivity improvements [7][12]. - A humorous take among programmers reveals their dissatisfaction with internal tools, indicating a struggle to adapt to less effective alternatives [9][12].