AISB is a 4-week long intensive program to bring researchers and engineers up to speed on security fundamentals for AI systems. Applications are open NOW for particpants, instructors, and mentors.
What will I learn?
Note: The program content is preliminary and may change.
Week 0: Prerequisites
- Linux Fundamentals: File permissions, process isolation, etc.
- Python: Optional coding exercises
Week 1: Security Fundamentals
- Cryptography & Cryptanalysis: Building and attacking encryption and hashing primitives
- Threat Modeling: Information security fundamentals, analyzing systems for vulnerabilities, security mindset
- Penetration Testing: Hands-on attacks on vulnerable systems, cyber kill chain phases, incident response
- Network Security: Building and breaking network and authentication protocols
Week 2: AI Infrastructure Security
- Process Isolation & Containerisation: Building Docker from scratch, namespace isolation, container escape techniques
- Supply Chain Security: Weak points in ML development and deployment environments, dependency provenance, defense best practices
- Application Security: XSS, SQL injection, buffer overflows, secure coding practices
- Security Operations: Hands-on building and maintaining cloud environments, monitoring, exploiting misconfigurations
Week 3: AI-Specific Security
- Model Training and Inference Basics in PyTorch
- Inference-Time Attacks Adversarial examples, prompt injection, exfiltration attacks
- Training-Time Attacks: Data poisoning, backdoors, checkpoint manipulation
- AI Application Security: Cross-user data exfiltration, MCP, RAG, injection attacks
- AI Infrastructure: GPU isolation and security features, ML pipeline security
Week 4: Capstone Project
- Choose your focus: Implement novel security solutions, replicate sophisticated attacks, or conduct authorized penetration testing
- Work with expert mentors on security challenges
- Present findings to cohort and industry professionals
Who should apply?
Our ideal candidate cares about safe and responsible development of AI systems. They also have some prior experience with deep learning (training/evals) and are comfortable with Python.
Some nice cherry-on-tops would include having a cybersecurity background, and C/C++ experience.
Candidates must be available to attend the program full time, in person in London.
FAQs
What is the application process like?
Fill out application → do a technical take-home → interview with someone on the team.
I am a high school or college student. Can I still apply?
We expect to have a mix of students as well as professionals, so you should apply if you fit the background criteria.
I have other commitments. Can I attend this program part time?
The curriculum and the readings will likely not leave much time for outside commitments, so we would recommend not having other major time commitments during the bootcamp.
Feel free to reach out if you'd like to discuss your specific circumstances.
What are the recommended prerequisites?
We recommend having a solid background in deep learning (or completing a program like MLAB or ARENA) and being comfortable with Python. These skills will help you make the most of the hands-on exercises and technical content during the bootcamp. We'll also send preparatory materials a couple of weeks before the program to help you brush up on essentials.
Can I join this program for some weeks but not others?
For participants: No, given we have limited spaces, we would be more excited to have participants who can commit to the entire four weeks of the program.
For mentors/instructors: Yes, you can participate for specific weeks. Please take a look at our syllabus to see which weeks align with your expertise.
What will an average day look like?
You'll start the day with a lecture or a deep dive into a significant vulnerability or exploit. After this, you will spend most of your time pair programming to work on the exercises, and reading relevant material. We will plan to wrap up the exercises by dinnertime, after which we'll have some additional reading material to prepare for the next day.
In the last week, you'll work on a capstone project with a mentor - this can be implementing a security solution, replicating a significant cyberattack (in a controlled environment), or trying to break into systems (again, with explicit permission)
Does this program cost money?
No. In fact, we will also take care of accommodation, travel, food (lunch/dinner on weekdays), and visas if needed.
I have more questions! Where can I reach out?
Shoot us an email!
Meet the Team
Jan Michelfeit
Security lead at Conjecture. Designs AISB's hands-on labs and capstone projects, drawing on 10+ years securing complex systems and ML infrastructure.
Pranav Gade
Research engineer at Conjecture. Created AISB to bridge AI safety and security, and leads curriculum design and program direction.
Jinglin Li
Software engineer and educator. Keeps AISB running smoothly.