News

Interesting links with a short summary. For long reads, see Articles.

  1. · via Latent.Space

    Researchers using OpenAI’s GPT-5.6 Sol and other AI models disclosed two PHP-engine memory-corruption vulnerabilities, CVE-2026-109618 and CVE-2026-109619, that allow unauthenticated exploitation against major CMSs. The coordinated response across the PHP ecosystem shows AI can accelerate security audits, but developers should monitor patches and update their PHP installations promptly.

    PHP core memory-corruption bugs (PHP2Shell) disclosed via AI-assisted researchlatent.space ↗
  2. · via Latent.Space

    Tenex pays AI engineers for delivered output rather than hours, aiming to align incentives as AI speeds up work. They keep story points for compensation but also consider non-ticket tasks like architecture design and customer context gathering. The model has scaled to a 75-person team serving Fortune 2000 and mid-market clients, highlighting how AI can reshape software development economics.

    Tenex experiments with output-based pay for AI-enhanced developmentlatent.space ↗
  3. · via Hacker News

    Mistral AI announced the launch of Voxtral 3, its latest language model. The new model aims to improve performance and efficiency over previous versions, which could benefit developers building AI-enhanced tools or services. Keeping an eye on such releases helps frontend engineers evaluate emerging AI capabilities for integration into web applications.

    Mistral releases Voxtral 3 language modelmistral.ai ↗
  4. · via Hacker News

    A recent blog post describes a tactic where authors publish fake statistics to get them referenced by large language models, leading to widespread citation across the web. The author notes that this can unintentionally amplify misinformation and affect downstream AI applications. Developers working with AI-generated content should be aware of how data sources are discovered and cited by LLMs.

    How publishing fabricated statistics can cause LLMs to cite your datapeec.ai ↗