LightNews LightNews
biosec-ai-project.bsky.social
@biosec-ai-project.bsky.social
3 followers 5 following 15 posts
Posts Replies Media Videos
biosec-ai-project.bsky.social
1/7
The intersection of AI and biosecurity demands our immediate attention. As AI systems become more powerful, they could dramatically lower barriers to designing biological agents. This isn't science fiction – it's a near-term challenge we need to address.
January 18, 2025 at 9:15 PM
biosec-ai-project.bsky.social
What if you could easily find and implement instructions on how to perform a terrorist act? Recent #LLM progress may lead to exactly that.
www.vox.com/future-perfe...
How AI could spark the next pandemic
New research highlights how language-generating AI models could make it easier to create dangerous germs.
www.vox.com
January 18, 2025 at 9:09 PM
biosec-ai-project.bsky.social
The next #pandemic can very well be supported by #AI. Multiple #threats are becoming more and more realistic.
#biosecurity #bio #safety
www.axios.com/2023/06/16/p...
January 18, 2025 at 9:01 PM
biosec-ai-project.bsky.social
Governmental #regulation is essential to achieve safe biological models functioning. We meed mandatory rules to prevent #pandemics.
doi.org/10.1126/scie...
AI and biosecurity: The need for governance
Governments should evaluate advanced models and if needed impose safety measures
doi.org
January 18, 2025 at 8:54 PM
biosec-ai-project.bsky.social
Using biological design tools can speed up both beneficial and dangerous activities. Complex set of measures is required for a good balance!

doi.org/10.48550/arX...
Towards Responsible Governance of Biological Design Tools
Recent advancements in generative machine learning have enabled rapid progress in biological design tools (BDTs) such as protein structure and sequence prediction models. The unprecedented predictive ...
doi.org
January 18, 2025 at 8:51 PM
biosec-ai-project.bsky.social
LLMs can already be applied to generate laboratory protocols. We need to prevent using these tools for hazardous activities!
#biosecurity
doi.org/10.48550/arX...
BioPlanner: Automatic Evaluation of LLMs on Protocol Planning in Biology
The ability to automatically generate accurate protocols for scientific experiments would represent a major step towards the automation of science. Large Language Models (LLMs) have impressive capabil...
doi.org
January 18, 2025 at 8:45 PM
biosec-ai-project.bsky.social
Meet EternaBrain: the tool that uses puzzle solving in a video game to train neural networks to improve RNA design
#science #RNA #dl #nn
doi.org/10.1371/jour...
EternaBrain: Automated RNA design through move sets and strategies from an Internet-scale RNA videogame
Author summary The design of RNA sequences that fold into target structures is a computationally difficult task whose importance continues to grow with the advent of RNA-based therapeutics and diagnos...
doi.org
January 18, 2025 at 8:42 PM
biosec-ai-project.bsky.social
Have you heard of RFdiffusion? This method, based on fine-tuning RoseTTAFold, leads to outstanding performance in protein design tasks! #AI
doi.org/10.1038/s415...
De novo design of protein structure and function with RFdiffusion - Nature
Fine-tuning the RoseTTAFold structure prediction network on protein structure denoising tasks yields a generative model for protein design that achieves outstanding performance on a wide range of prot...
doi.org
January 18, 2025 at 8:25 PM
biosec-ai-project.bsky.social
GPT-4 simplifies laboratory automation: simple prompts can generate expert-level performing scripts with contextual understanding #AI #GPT4
doi.org/10.48550/arX...
LLMs can generate robotic scripts from goal-oriented instructions in biological laboratory automation
The use of laboratory automation by all researchers may substantially accelerate scientific activities by humans, including those in the life sciences. However, computer programs to operate robots sho...
doi.org
January 18, 2025 at 7:38 PM