Researchers have found that LLM-driven bug finding is not a drop-in replacement for mature static analysis pipelines. Studies comparing AI coding agents to human developers show that while AI can be ...

In a preview stage, Code Review launches a team of agents that look for bugs in parallel, verify them to filter out false positives, and rank them by severity.

Microsoft's February 2026 Foundry update includes broader platform changes, but the most immediate developer-facing news for VS Code users is an AI Toolkit refresh centered on tool discovery, agent ...

COBOL was associated with the Y2K phenomenon at the turn of the century, a software problem arising from the inability of ...

研究团队表示，三款模型基于相同的基础训练数据集，高一致率的结果符合预期。真正具备研究价值的是模型间25%的分歧部分，这种差异大概率并非源于模型对工具质量的独立判断，而是由基于人类反馈的强化学习（RLHF）调优策略不同，以及生成环节的专属微调差异导致。

有趣的是，Claude Code在不同项目上下文中的表现也颇具特色。尽管同一工具类别在不同代码仓库中，其选择可能会有所不同，但在相同项目中，即使用不同的措辞表达需求，其选择的稳定性平均达到76%。这表明，项目的上下文对工具选择的影响远大于指令的措辞。 从实验结果来看，Claude ...

Chainguard, the trusted source for open source, today announced it has expanded Chainguard Libraries coverage across Python, Java, and JavaScript, with customers seeing 94% coverage across the Python ...

While Anthropic’s Claude Code grabbed headlines, IBM has been deploying its own generative AI solution, Watsonx Code Assistant for Z, designed to modernize the very mainframes it built. Unlike general ...

On3, 2026, Anthropic unveiled Claude Code, an AI tool capable of modernizing decades-old COBOL systems in months rather than years.

Anthropic's Claude Code modernising COBOL systems triggered a 13.2% correction on IBM's stock.

A blog post by Anthropic pointing out that Claude Code can refactor legacy COBOL applications appears to have sent IBM’s share price down by double ...

Investors reacted to Anthropic’s assertion that Claude Code could streamline legacy COBOL modernization, raising concerns about pressure on IBM’s high‑margin mainframe services despite longstanding ...