Viqus Logo Viqus Logo
Home
Categories
Language Models Generative Imagery Hardware & Chips Business & Funding Ethics & Society Science & Robotics
Resources
AI Glossary Academy CLI Tool Labs
About Contact

Laude Institute Launches ‘Slingshot’ AI Grants Program

Artificial Intelligence AI Grants Tech Research Laude Institute AI Evaluation Startup Funding TechCrunch
November 06, 2025
Viqus Verdict Logo Viqus Verdict Logo 8
Progress, Not a Paradigm Shift
Media Hype 6/10
Real Impact 8/10

Article Summary

The Laude Institute’s new ‘Slingshot’ program represents a significant injection of resources into the AI research landscape. Aimed at accelerating progress, the grants offer funding, compute power, and engineering support to fifteen diverse projects, with a key focus on the notoriously difficult task of AI evaluation. Projects include established benchmarks like Terminal Bench and ARC-AGI, alongside innovative approaches from Formula Code and BizBench. The program’s emphasis on dynamic competition through CodeClash, led by John Boda Yang, highlights a shift towards more rigorous assessment methods. This initiative reflects a growing concern about reliance on company-specific benchmarks and aims to foster a more robust and adaptable evaluation ecosystem. The program's selection of established researchers and new, disruptive ideas demonstrates a strategic effort to address critical shortcomings in the field.

Key Points

  • The Laude Institute is providing substantial resources to accelerate AI research.
  • A primary focus is on improving the difficult problem of AI evaluation.
  • The program incorporates established benchmarks alongside novel approaches like CodeClash.

Why It Matters

This news is important because it signifies a move towards more practical and competitive AI evaluation. The shift away from company-specific benchmarks is crucial for ensuring the long-term reliability and generalizability of AI systems. Funding such research will not only speed up progress but also potentially mitigate risks associated with blindly trusting AI outputs based on proprietary data. This initiative is relevant for anyone involved in developing, deploying, or regulating AI, as it directly addresses a fundamental challenge within the field.

You might also be interested in