The AIxBio Hackathon, co-organized with BlueDot Impact brings together researchers, engineers, biosecurity professionals, and AI safety enthusiasts to work on one of the most urgent intersections in safety: how AI is changing biological risk, and what we can build to stay ahead.

Over three days, participants will develop tools, prototypes, and research addressing real gaps in biosecurity infrastructure, from DNA synthesis screening and pandemic early warning systems to practitioner tools that don't exist yet.

Top teams get

*Not all winning projects may receive a fellowship invitation. An invitation might also depend on other external factors like the Apart Core team's capacity to onboard new research teams.

What is AIxBio?

AI is transforming biology faster than biosecurity can keep up. Open-weight protein design models like Evo2 (40B parameters, trained on 128,000+ genomes) and tools like RFdiffusion are making it possible to design novel biological sequences with less expertise than ever before. At the same time, benchtop DNA synthesizers are approaching the capability to print virus-length sequences (expected within 2-5 years), and current screening infrastructure covers only a fraction of global synthesis capacity.

On the defensive side, pathogen-agnostic metagenomic surveillance is finally becoming practical. Wastewater monitoring can detect outbreaks weeks before hospitals see cases. AI anomaly detection could identify engineered pathogens in sequencing data. But most of these tools are siloed, under-resourced, or still in the research phase. The field needs builders.

Hackathon Tracks

1. DNA Synthesis Screening and Biosecurity Guardrails

Build better tools for screening DNA synthesis orders, detecting dangerous sequences, and creating guardrails for AI-powered biological design tools. Current screening misses AI-designed protein variants, struggles with short sequences, and doesn't cover benchtop synthesizers at all.

2. Pandemic Early Warning Systems

Develop AI-powered surveillance, anomaly detection, and data integration tools for catching outbreaks earlier. Wastewater monitoring, metagenomic sequencing, and open-source disease intelligence are all ripe for better tooling.

3. Biosecurity Infrastructure and Practitioner Tools

Build the tools that biosecurity practitioners actually need: unified dashboards, rapid risk assessment systems, policy trackers, secure communication tools, and accessible resources for under-resourced institutions.

Who should participate?

• AI safety researchers and engineers

• Biosecurity and public health professionals

• Machine learning researchers interested in biological applications

• Security researchers and red teamers

• Policy researchers working on biogovernance

• Software engineers interested in building safety infrastructure

• Students and early-career researchers exploring biosecurity

No prior biosecurity experience is required. The Resources tab has a curated reading list organized by track. Teams typically form at the start of the event.

What you will do

Over three days, you will:

1. Form teams and choose a challenge track

2. Research and scope a specific problem using the provided resources

3. Build a project: a tool, prototype, evaluation, or research contribution

4. Submit a research report (PDF) documenting your approach, results, and implications

5. Present your work to judges from leading biosecurity and AI safety organizations

What happens next

After the hackathon, all submitted projects will be reviewed by expert judges. Top projects receive prizes and are featured on the Apart Research website. Outstanding work may be invited for further development, publication, or presentation at upcoming events.

Why join?

Apart Research has organized 55+ research sprints with 6,000+ participants across 200+ global locations. Co-organized with BlueDot Impact. Our hackathons produce real research output: published papers, new research collaborations, and contributions to open-source safety tools used across the field.