Likely AI-generated. Pangram flags 97% of this text as AI-written.
2
Hypervisor-Zero: Swarm Agents for AI Cyber Defense and Control
0
ProposalGrant
Closes November 22nd, 2026$0raised
$1,000minimum funding
$20,000funding goal
Offer to donate
31 daysleft to contribute
You're pledging to donate if the project hits its minimum goal and gets approved. If not, your funds will be returned.
$20,000 left to reach the funding goal.
Project summary
Project public demo:
Hypervisor-Zero is building a swarm-agent AI cyber defense layer where specialized agents work together to detect threats, investigate attacks, and coordinate defensive responses. One agent monitors system activity, another analyzes suspicious behavior, others investigate vulnerabilities and validate findings, while a coordinating agent prioritizes threats and recommends containment actions, So that the division of labor enables faster cross-validation, broader threat coverage, and fewer false positives than relying on a single agent. A central control layer enforces permissions and security policies, while isolated execution and shadow environments allow proposed responses to be tested before deployment.
What are this project's goals? How will you achieve them?
The goal is to deliver a working prototype and practical research demonstrating how coordinated, controlled AI agents can improve cyber defense
How will you know it worked?
Success will be measured by threat detection accuracy, false positive rates, time to identify and contain threats, and the system’s ability to block unauthorized agent actions even when agents are compromised or manipulated. We will compare results against baseline systems and publish documented findings, including failures and limitations, so others can independently assess the results.
How will this funding be used?
Expense| Budget $20,000
H200 GPU compute: Approximately 1,100 H200 GPU-hours at $5.29/hour for model inference, fine-tuning, and evaluation runs.| $6,000
LLM API usage: OpenAI, Anthropic, and other model providers for swarm coordination, threat analysis, verification, and adversarial testing.| $4,500
CPU compute and isolated sandboxes: Agent execution, secure environments, and simulated cyberattack testing.| $3,000
Application hosting and backend infrastructure: Dashboard hosting, backend services, and managed databases.| $1,500
Storage and backups: Experiment traces, security logs, datasets, and backups.| $800
Independent security assessment: Testing the dashboard, APIs, access controls, and agent-enforcement mechanisms.| $2,500
Testing and operational infrastructure: Automated testing, monitoring, job queues, and network-related costs.| $1,700
Total| $20,000
Who is on your team? What's your track record on similar projects?
I am (Muhammad) a third-year software engineering student and independent researcher I'm currently solo and I previous AI safety work focused on developing agricultural datasets and fine-tuning models to understand agricultural knowledge and local language's
Project failure
Failure could mean the swarm misses real attacks, generates excessive false alarms, or takes too long to respond. Compromised agents might bypass controls or trigger unsafe defensive actions.
How much money have you raised in the last 12 months, and from where?
I self funded my work using personal funds and limited resources I have not raised significant external funding yet!