cd /news/ai-safety/if-you-re-agentic-work-in-biosecurit… · home topics ai-safety article
[ARTICLE · art-64775] src=forum.effectivealtruism.org ↗ pub= topic=ai-safety verified=true sentiment=↑ positive

If you're agentic, work in biosecurity

A call to action urges agentic individuals to work in biosecurity, arguing that the field is neglected, tractable, and ready for high-execution talent without requiring a biology background. The post highlights ecosystem growth and unmet needs, offering personal pitches to potential recruits.

read8 min views1 publishedJul 18, 2026

Published on July 18, 2026 5:38 PM GMT

*Disclaimer: Although I work on the Groups Team at CEA, I’m writing this in a personal capacity, and this post does not constitute an endorsement by CEA. *

**Agency - the realisation that you ****really can **just do things.

Biosecurity needs people (of any background) who are agentic and have a high execution velocity and track record. The field is ready to absorb individuals who feel a heroic responsibility to build teams and start organisations that own key ecosystem gaps. You do not need a background in biology to be a great biosecurity operator. For every stage of your career, there's an unmet need in the ecosystem you can directly address. If you're still unsure, I'll personally pitch you on why and how to best help.

There's been extensive writing on the case for working on existential risks and biosecurity. I signpost some readings here and here that describe this better and more succinctly than I can. I want to focus on why now is the time to commit to biorisk. It's estimated that around ~160 people are working full-time on reducing the near-infinite ways in which mirror bacteria or engineered pandemics could pose existential risk to humanity's future and flourishing. Despite this, it's estimated that almost 95% of this risk can be reduced by deploying relatively tractable interventions: PPE, biohardening, detection, and medical countermeasures (AIxBio is worthy of its own post). It's incredibly tractable, neglected, and high-impact to work in a field where there's such strong consensus on how to reduce risk. The primary constraint at the moment is people.

There are strong reasons to be excited about biosecurity fieldbuilding right now. Cambridge Biosecurity Hub has had success growing the community and becoming an owner for stewarding the movement; Bluedot is trialling the Voyager Fellowship to locate much-needed entrepreneurial talent after relaunching the Biosecurity Course; Pivotal has launched a Biodefense Fellowship; CG and Bluedot have begun collaborating on rapid and career transition grants. Personally, many of my collaborators and friends have transitioned into direct work in both the GCBR and meta-space, where the possibilities for introducing biosecurity to talented people are growing.

AI Safety has recently had a generalist resurgence that biosecurity could draw learnings from in how we characterise talent gaps. There has also been exciting growth within EA/AIS ecosystem building (e.g. Kairos' SPAR and CBAI starting AIxBio fellowships/streams; the launch of Impact First; NEST, and growth within CEA) that I'm excited to see more collaboration and cross-learning between. Few spaces are as welcoming of entrepreneurial and fast-shipping talent as biosecurity right now.

I spend most of my time at CEA liaising with students looking to transition into direct GCR work. Although the GCBR community tends to be geared towards mid-career/senior professionals, there are still ways to contribute whilst upskilling. Students are in an enviable position of limitless time and energy, which we should all be leveraging more.

  • Hosting events
  • Full-time staff have limited capacity to do this. Hosting happy hours, offering support to those trying to arrange summits, and working with your universities to host convenings with academics in adjacent fields (e.g. metagenomics, supply chains, etc.) can be unusually high-tailed.

Launch a biosecurity group on your campus (especially if you're at a top university)- EA and AIS have grown considerably due to the bottom-up efforts of local groups. Students (from undergraduates to postdocs) who lead effective biosecurity groups can provide an early-career funnel for organisations looking to hire junior talent, and most importantly, make biorisk one of the top issues students and graduates want to work on. A single group could steer dozens of students into the field.

  • I've become increasingly sceptical of concerns about the delayed impact of students. I think there's far more to do than just direct technical work, and generally increasing the worldwide awareness of biorisk seems counterfactually worthwhile. At the same time, student AIS groups such as OASI, MAIA, and AISST have provided a model for student involvement towards GCRs I'd be curious to explore in biosecurity.

  • Generally speaking, growing your local EA group can increase the peripheral knowledge about GCBRs in your school and the agency of students to pursue an impact-aligned career.

  • Encourage career transitions

  • Students often overlook this. Agentic community builders (especially students) have privileged access to coffee- and career-chats with those later on in their careers. Biosecurity needs individuals with a broad skillset across a range of fields (from operations, logistics, entrepreneurs, civil servants). Pitch people on why they should work on biorisk, and encourage them to do a BlueDot course or fellowship. A counterfactual career-career transition contributes thousands of person-hours of additional capacity if sufficiently scoped.

  • Experiment with media communication

  • I want to caution that, in this context, I encourage students to write for their college papers and outlets about:

  • Topics they feel are neglected and that people should work on (e.g., research narratives about

[mirror bacteria](https://www.mbdialogues.org/)) - Fellowships or
[opportunities](https://gcbrupdates.substack.com/) in GCBRs - Encouraging students to be agentic and avoid
[common failure modes](https://www.ft.com/content/f01c0d6d-9546-4c3d-b1f0-18adc301ce11) of high-performing talent

Although I have less interface with mid-career professionals, there are a multitude of ways to get involved in biosecurity right now. The ecosystem needs people with expertise that is useful__ __to GCBRs, and it's widely understood that individuals can pick up the necessary context. We need creative problem solvers who are ready to use their background towards one of the four pillars in biosecurity.

Skills I'm especially excited about include:

  • Field-builders (of all stages) looking to collaborate on ecosystem growth
  • People who have experience working with the government, national security, or policy
  • People who have ops/chief of staff/founder skillsets
  • People who have worked in supply chains, logistics, or distribution are taking on PPE stockpiling
  • Seeing the more traditional AIS crowd consider taking on AIxBio challenges

__I list below a first-pass list of ways to get involved. Much of this is also captured on https://biosecurity.world: __ Courses

Fellowships

Projects to own

Places to build

RFPs and funders

To follow

While most fieldbuilding discourse is centred around vertical movement through the ecosystem, I'm really excited to see increased communication between fieldbuilders and those working in direct (technical or policy) work. One thing I've noticed about AIS is the shared sense of responsibility towards ecosystem growth by those working in organisations of all sizes. As a fieldbuilder, I'd encourage those currently working to consider where their comparative advantage is in contributing to the mission.

  • Encouraging more thought-leadership/public-facing writing in your organisation

  • Notwithstanding the sensitive and infohazardous nature of a significant portion of GCBR work, writing shallow articles on new focus areas, topics you think should be worked on, or where you would like to see the field can help fieldbuilders better prioritise our referrals, connections, and initial conversations. I've found the SecureBio and Defenses in Depth substacks particularly useful. I do believe the EA Forum is still a strong place for the community to convene and exchange ideas.- This could come in various forms, from posting on Linkedin, writing on Substack, or engaging in forum-posting.

  • Mentor (early-career) talent

  • Having launched and scaled almost 3 mentored-research programs thus far, the biggest constraint for any program is mentor availability. The value-add of your time is highly asymmetric - even 20-30 minutes a week could be the deciding factor in a counterfactual career transition. Many mentors I have worked with tend not to consider this as an option until asked; I strongly encourage anyone (at any stage) to consider becoming a SPAR mentor, support a fellowship, or even post a project on the forum seeking mentees. Effective Thesis has a bank of research topics that's routinely updated to draw and contribute to.

  • Launch your own programs

  • Consider whether it makes sense to launch your own research, fellowship, or entrepreneurship programs. It's undeniable that there is a staff cost to this, but the upside of well-scoped programming that meets specific needs could be huge. Take the example of AIxBio and the successes of fellowships in directing the talent pool towards this particular area. I'm keen to speak to orgs considering their own fellowship programs/help soundboard.

  • An alternative here is collaborating to start local biosecurity hubs, a shout-out to Li-Lian Ang, who

[wrote about this](https://lilianang.substack.com/p/we-need-more-local-ai-safety-hubs) for AIS.

- Tap into the meta-space
  • The meta-space has unrealised potential and reflexivity to wider EA/GCR cause prioritisation. I'd be excited to see those working in biosecurity tap more into the institutional knowledge of GCR and EA capacity builders. In our roles, we're constantly strategising about community growth, funnel conversion, and how to increase the pool of talented (EA- or impact-aligned) individuals. Growing biosecurity is as much of a cause-agnostic field-building challenge as it is a biosecurity-specific one.
  • I'm fairly convinced that there's a greater institutional framework for developing and directing talent than the community may appreciate. I suspect the limiting factor here is a wider coordinated fieldbuilding effort/strategy.

(Additional) Disclaimer

  • Many of these ideas demand far more nuance than I've given. I strongly encourage readers to red-team these ideas and treat them as prompts for your own reflection and discussion.

I'd be really excited to see a forum discussion on:

  • Ways to better coordinate as a field
  • What thought-leadership in biosecurity fieldbuilding should look like
  • Exciting fieldbuilding projects people are considering or looking to collaborate on

Discuss

── more in #ai-safety 4 stories · sorted by recency
── more on @cea 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/if-you-re-agentic-wo…] indexed:0 read:8min 2026-07-18 ·