Vals AI Blog
Research, benchmark deep dives, and evaluation findings from the Vals AI team.
- Posts
- 26 from 26 Nov 2025 to 28 Oct 2026
- Posting
- weekly (26 in the last year)
- Kind
- company blog
- Feed
- https://www.vals.ai/blogs/rss.xml
2026 25
- Can AI Get Immigration Right?
- Why is my local Qwen sending requests to Alibaba Cloud?
- Thomson, One N at a Time
- Evaluating Teen Safety
- How Far Will AI Assistants Go for a Refund They Can’t Get?
- Routers Are a New AI Primitive
- How pro-China are Chinese models?
- Do Agent Teams Pay Off? A Case Study on Vibe Code Bench
- An Evaluation of Inception's Mercury Decide
- Trust but Verify: Our first Public Sector Summit
- Two-Thirds of MiMo v2.6's Coding Tasks Leak the Answer. Are Models That Exploit This Misbehaving?
- An Independent Evaluation of TypeSafe's Jev
- Two Room-Temperature Antiferromagnetic Semiconductor Candidates
- Vals Web Search Index: Evaluating Search for Real Work
- Has the Bitter Lesson Come for AI Detectors?
- A Lean Proof of the Thomson Problem for Seven Electrons
- Evaluating AI Safety in Teen Conversations
- A Faster Shortest Path Algorithm
- AI Cheating is on the Rise
- Vals Environmental Impacts Report
- Claude Fable 5.1 Solves the Cyphral Distich
- Series A: Always a Higher Peak
- Child AI Safety
- Vals Public Sector Memo
- Measuring Healthcare AI Where It Actually Works