# SPAR Spring 2026: Expanding the Pipeline for AI Safety Research

> Coverage of lessw-blog

**Published:** December 20, 2025
**Author:** PSEEDR Editorial
**Category:** risk
**Content tier:** free
**Accessible for free:** true



**Word count:** 365


**Tags:** AI Safety, Research Fellowship, SPAR, Biosecurity, AI Governance, Mentorship

**Canonical URL:** https://pseedr.com/risk/spar-spring-2026-expanding-the-pipeline-for-ai-safety-research

---

The Supervised Program for Alignment Research (SPAR) announces its largest cohort yet, offering over 130 projects and a new biosecurity track to aspiring researchers.

In a recent announcement on **lessw-blog**, the team behind the Supervised Program for Alignment Research (SPAR) opened applications for their Spring 2026 cycle. This program has established itself as a critical entry point for talent in the AI risk ecosystem, and this upcoming round represents a significant scaling of their efforts. With over 130 distinct research projects available, this is the largest fellowship the organization has offered to date.

**Contextualizing the Opportunity**  
As frontier models continue to advance, the demand for skilled researchers capable of addressing alignment, governance, and security risks outpaces the current supply. The primary bottleneck is often access: aspiring researchers struggle to find structured guidance or entry-level projects that allow them to build a portfolio. SPAR addresses this by pairing mentees with experienced mentors in a remote, part-time format (5-10 hours per week). This structure lowers the barrier to entry, allowing students and professionals to contribute to high-impact work without relocating or leaving their current positions.

**Key Developments in Spring 2026**  
The source highlights a notable expansion in the program's scope. Beyond the traditional focus on technical alignment and governance, this cycle introduces a dedicated **Biosecurity** track for the first time. The distribution of projects reflects the multifaceted nature of modern AI risk:

*   **Alignment, Evals & Control:** ~58 projects
*   **Policy & Governance:** ~45 projects
*   **Security:** ~21 projects

The program is scheduled to run from February 16 to May 16. While the final deadline for applications is January 14, the post emphasizes that mentor reviews occur on a rolling basis, heavily incentivizing early submission.

**Why It Matters**  
For the broader community, this announcement signals a maturing infrastructure for AI safety research. The ability to support 130+ simultaneous projects suggests a growing capacity for mentorship and organized research management within the field. For individuals, it represents a concrete pathway to transition into AI safety work.

We encourage readers to explore the full list of projects to understand the specific research questions currently being prioritized by mentors.

[Read the full post at lessw-blog](https://www.lesswrong.com/posts/AacKhcv3r3DKybyMB/spar-spring-2026-130-research-projects-now-accepting)

### Key Takeaways

*   SPAR Spring 2026 is the largest cohort to date, featuring over 130 research projects.
*   The program has expanded to include a dedicated Biosecurity track alongside Alignment, Governance, and Security.
*   The fellowship is remote and part-time (5-10 hours/week), running from February 16 to May 16.
*   Applications are open until January 14, but rolling admissions favor early applicants.

[Read the original post at lessw-blog](https://www.lesswrong.com/posts/AacKhcv3r3DKybyMB/spar-spring-2026-130-research-projects-now-accepting)

---

## Sources

- https://www.lesswrong.com/posts/AacKhcv3r3DKybyMB/spar-spring-2026-130-research-projects-now-accepting