Who we serve

DARPA

At Aptima, we’re advancing the frontiers of national security by developing breakthrough technologies for DARPA that transform how humans and machines work together to achieve decision superiority. Our research pushes the limits of Warfighter training, performance, and decision-making, integrating operators with cutting-edge systems to create capabilities that are faster, smarter, more autonomous, and more resilient than ever before.

Explore our solutions created in partnership with DARPA

Our decision advantage systems help analysts and operators make faster, more informed decisions in high-pressure, high-uncertainty environments.

Accelerate biodefense and biosecurity insight with BIOSAGE.

Our solution gives intel analysts AI-powered cross-domain knowledge discovery and synthesis tools that sharpen decision advantage. BioSage breaks down stovepipes, reduces analysis time, and delivers faster, clearer understanding to inform better decisions across Army missions.

 

Biosage Papers:

  • BioSage: Human-Agent Collaboration Platform for Cross-Disciplinary Knowledge Discovery and Synthesis. Klinefelter, Z., Erickson, I., …, & Volkova, S. Proceedings of the Human Factors and Ergonomics Society Annual Meeting 2025.
  • Cross-Disciplinary Knowledge Retrieval and Synthesis: A Compound AI Architecture for Scientific Discovery. Volkova, S., Bautista, P., Hiriyanna, A., Ganberg, G., Erickson, I., Klinefelter, Z., Abele, N., Kao, H.-T., & Engberson, G. (2025) (under review).

Our modeling, simulation, and wargaming technologies help DARPA programs explore complex futures, test high-risk ideas safely, and develop strategic decision advantage.

Strengthen human-agent team coordination, planning, and performance with SPIDERSENSE and EMHAT.

SPIDERSENSE is an AI-enabled planning and team-optimization testbed that transforms team behavioral and mission data into machine-readable plans and intelligent recommendations. EMHAT is a human-agent modeling and evaluation platform that enables simulated experimentations with human and agent attributes to achieve optimal human-agent team performance. These solutions foster shared situational awareness, accelerate decision-making, and optimize team performance across planning, execution, and rapid reconstitution.

 

EMHAT and SPIDERSENSE papers:

  • VirTLab-Eval: Human-AI Team and Human Digital Twin Performance Evaluation. Volkova, S., Kao, H., Penafiel, L., Nguyen, D., Cohen et al. Proceedings of the Human Factors and Ergonomics Society Annual Meeting 2025.
  • VirtLab: An AI-Powered System for Flexible, Customizable, and Large-scale Team Simulations. Almutairi, M., Chiang, C., …, S. Volkova, & Gomez-Zara, D. (2025). UIST System Demonstrations.
  • Exploring Big Five Personality and AI Capability Effects in LLM-Simulated Negotiation Dialogues. Cohen, M., …, & Volkova, S. (2025). KDD Workshop on Agentic and Generative AI Evaluation.
  • VirtLab: Augmented Intelligence for Training and Evaluating Human-AI Team Dynamics through Digital Twin Interactions. S. Volkova, D. Nguyen, L. Penafiel et al. HCI International. 2025.
  • Exploratory Models of Human-AI Teams: Human Digital Twins to Investigate Trust Development. Nguyen, D., Cohen, M., …, Volkova, S. In Proceedings of the Multidisciplinary Perspectives on Human-AI Team Trust. 2025.

Our cognitive warfare technologies help DARPA assess, model, and proactively build resiliency against adversary influence and information manipulation.

Give MISO and PSYOP operators and planners a decisive advantage with the BRIES compound-AI system.

Our solution enables faster insight, proactive counter-narrative strategies, and optimized Soldier–AI performance. The systems use autonomous agents and validated human digital twins of individuals, communities and populations to test counterfactual scenarios and strategic messaging pre-deployment to simulate influence, information resilience, and optimize human-AI teaming dynamics.

 

BRIES papers:

  • Community-Aligned Behavior Under Uncertainty: Evidence of Epistemic Stance Transfer in LLMs. Gerard, P., Chang, A., & Volkova, S. EurIPS 2025 Workshop on Epistemic Intelligence in Machine Learning.
  • Modeling Information Narrative Detection and Evolution on Telegram during the Russia-Ukraine War.Gerard, P., Volkova, S., Penafiel, L., Lerman, K., & Weninger, T. Proceedings of the International AAAI Conference on Web and Social Media (ICWSM) 2025, Vol. 19, pp. 602-614.
  • Compound AI for Decision Advantage: Human Digital Twin Agent Teams in Military Operations. S. Volkova, L. Cassani et al. 17th International MODSIM World Conference 2025.
  • SOTOPIA-S4: A User-Friendly System for Flexible, Customizable, and Large-Scale Social Simulation.X. Zhou, Z. Su, S. Feng, …, S. Volkova, et al. (2025) NAACL System Demonstrations.
  • ValueScope: Unveiling Implicit Norms and Values via Return Potential Model of Social Interactions.** CY. Park, S. Li, H. Jung, S. Volkova, T. Mitra, D. Jurgens, Y. Tsvetkov. EMNLP 2024.
  • Proactive Defense: Compound AI for Detecting Persuasion Attacks and Measuring Inoculation Effectiveness. Volkova, S., Dupree, W., Kao, H.-T., Bautista, P., Ganberg, G., Beaubien, J., & Cassani, L. (2025). (under review).
  • Building Resilient Information Ecosystems: Large LLM-Generated Dataset of Persuasion Attacks. Kao, H.-T., Panasyuk, A., Bautista, P., Dupree, W., Ganberg, G., Beaubien, J. M., Cassani, L., & Volkova, S. (2025). (under review).

Get on-contract in record time,
with our approved contract vehicles

We have a portfolio of government contract vehicles that allow us to quickly and efficiently innovate solutions to your most pressing needs.

Learn more about our contract vehicles

Equip your team with the future of defense innovation.

Partner with Aptima today.