Generative AI is transforming the software industry, with tools like Claude Code, GitHub Copilot or others, enabling developers to achieve unprecedented levels of efficiency. While this is exciting progress, it also raises important concerns, encouraging stakeholders to approach these technologies with care. Current AI tools often lack the nuanced understanding necessary to address subtle, yet critical aspects of software development, particularly in the domain of security.
This course provides a comprehensive insight into the responsible use of generative AI in coding. Participants delve into topics in software development that are most likely to be impacted by careless use of generative AI, including authentication, authorization, and cryptography. The curriculum also includes an analysis of how Claude Code, GitHub Copilot or others handle secure coding practices related to key vulnerabilities outlined in the OWASP Top Ten, such as path traversal, SQL injection, or cross-site scripting.
Through hands-on learning and experimenting, participants will get a solid understanding of both the strengths and limitations of AI-assisted development. In addition, case studies of real-world incidents showcase the consequences of insecure code and demonstrate the dual nature of generative AI as both a resource and a potential risk.
By the end of the course, developers will be equipped with the knowledge and skills to integrate AI tools into the software development lifecycle responsibly, enhancing efficiency without compromising security or product quality.
This variant of the Python security course deals extensively with how certain security problems in code are handled by GenAI. Through a number of hands-on labs participants will get first-hand experience about how to use GenAI responsibly, and how to prompt it to generate the most secure code. In some cases it is trivial, but in most cases it is not; and in yet some other cases it is basically impossible. At the same time, the labs provide general experience with using GenAI in everyday coding practice - what you can expect from it, and what are those areas where you shouldn't rely on it.
PRACTICAL INFO
Understanding the essentials of responsible AI
Getting familiar with essential cyber security concepts
Understanding Web application security issues
Detailed analysis of the OWASP Top Ten elements
Putting Web application security in the context of Python
Going beyond the low hanging fruits
Managing vulnerabilities in third party components
Python developers using GenAI tools.
Preparedness:
General Python and Web development.
Day 1
Coding responsibly with GenAI
What is security?
Threat and risk
Cyber security threat types – the CIA triad
Consequences of insecure software
What is responsible AI?
Security and responsible AI in software development
The OWASP Top Ten 2025
A01 - Broken Access Control
· Access control basics
· Case study – Broken authz in FIFA platform during 2026 World Cup
· Confused deputy
o Insecure direct object reference (IDOR)
o Path traversal
o Lab – Insecure Direct Object Reference
o Path traversal best practices
o Lab – Experimenting with path traversal in GenAI
o Authorization bypass through user-controlled keys
o Case study – Remote takeover of Nexx garage doors and alarms
o Lab – Horizontal authorization
o Unrestricted file upload
o Good practices
o Lab – Unrestricted file upload
· Server-side Request Forgery (SSRF)
o Case study – SSRF in Ivanti Connect Secure
A02 - Security Misconfiguration
· Configuration principles
· Python configuration best practices
o Configuring Flask
· Web security configuration issues
o Content Security Policy
o Fetch directives
o Source allowlisting
o Strict CSP: using nonces and hashes
o CSP best practices
· Cookie security
o Cookie attributes
· Secrets management
o Hard coded passwords
o Best practices
o Lab – Hardcoded password
· XML entities
o DTD and the entities
o Entity expansion
o External Entity Attack (XXE)
- File inclusion with external entities
- Server-Side Request Forgery with external entities
- Lab – External entity attack
- Preventing XXE
- Lab – Prohibiting DTD
- Case study – XXE vulnerability in Ivanti products
- Lab – Experimenting with XXE in GenAI
Day 2
The OWASP Top Ten 2025
A03 - Software Supply Chain Failures
· Using vulnerable components
· Assessing the environment
· Hardening
· Untrusted functionality import
· Malicious packages in Python
· Supply chain security and the Software Bill of Materials (SBOM)
· SBOM examples
· Case study – The Polyfill.io supply chain attack
· Vulnerability management
o Patch management
o Vulnerability management
o Vulnerability databases
o Lab – Finding vulnerabilities in third-party components
o DevOps, the CI / CD build process and Software Composition Analysis
o Dependency checking in Python
o Lab – Detecting vulnerable components
· Security of AI generated code
o Practical attacks against code generation tools
o Dependency hallucination via generative AI
o Case study – A history of GitHub Copilot weaknesses (up to mid-2025)
o Case study – Agentic coding and code security (mid-2025 onward)
A05 - Injection
· Input validation
o Input validation principles
o Denylists and allowlists
o Case study – Denylist failure in urllib.parse.urlparse()
o What to validate – the attack surface
o Where to validate – defense in depth
o When to validate – validation vs transformations
· SQL injection
o SQL injection basics
o Lab – SQL injection
o Attack techniques
o Content-based blind SQL injection
o Time-based blind SQL injection
o SQL injection best practices
- Input validation
- Parameterized queries
- Lab – Using prepared statements
- Case study – SQL injection against US airport security
· Code injection
o Code injection via input()
o OS command injection
- Lab – Command injection
- OS command injection best practices
- Avoiding command injection with the right APIs
- Lab – Command injection best practices
- Lab – Experimenting with command injection in GenAI
- Case study – Shellshock
- Lab - Shellshock
- Case study – Command injection in Ivanti security appliances
· HTML injection - Cross-site scripting (XSS)
o Cross-site scripting basics
o Cross-site scripting types
- Persistent cross-site scripting
- Reflected cross-site scripting
- Client-side (DOM-based) cross-site scripting
o Lab – Stored XSS
o Lab – Reflected XSS
o Case study – XSS to RCE in Teltonika routers
o XSS protection best practices
- Protection principles - escaping
- Lab – XSS fix / stored
- Lab – XSS fix / reflected
- Case study – XSS vulnerabilities in DrayTek Vigor routers
Day 3
The OWASP Top Ten 2025
A06 - Insecure Design
· The STRIDE model of threats
· Secure design principles of Saltzer and Schroeder
o Economy of mechanism
o Fail-safe defaults
o Complete mediation
o Open design
o Separation of privilege
o Least privilege
o Least common mechanism
o Psychological acceptability
· Client-side security
o Frame sandboxing
- Cross-Frame Scripting (XFS) attacks
- Lab – Clickjacking
- Clickjacking protection best practices
- Lab – Using CSP to prevent clickjacking
A07 - Authentication Failures
· Authentication
o Authentication basics
o Multi-factor authentication (MFA)
o Case study – The InfinityGauntlet attack
· Password management
o Storing account passwords
o Password in transit
o Lab – Is just hashing passwords enough?
o Dictionary attacks and brute forcing
o Salting
o Adaptive hash functions for password storage
o Lab – Using adaptive hash functions in Python
o Lab – Experimenting with adaptive hash functions in GenAI
· Password policy
o NIST authenticator requirements for memorized secrets
o Password database migration
A08 - Software and Data Integrity Failures
· Subresource integrity
o Importing JavaScript
o Lab – Importing JavaScript
o Case study – The British Airways data breach
· Insecure deserialization
o Serialization and deserialization challenges
o Integrity – deserializing untrusted streams
o Deserialization with pickle
o Lab – Deserializing with Pickle
o Case study – The security of the machine learning supply chain
o Case study – The first wave of supply chain attacks: RCE via pickle (2022)
o Case study – Compromising the Hugging Face Hub repository
o Integrity – deserialization best practices
A10 - Mishandling of Exceptional Conditions
· Error and exception handling principles
· Error handling
o Returning a misleading status code
o Information exposure through error reporting
- Information leakage via error pages
- Lab – Flask information leakage
- Case study – Information leakage via errors in Apache Superset
· Exception handling
o In the except block. And now what?
o Empty except block
o Lab – Exception handling mess
Wrap up
Secure coding principles
· Principles of robust programming by Matt Bishop
And now what?
· Software security sources and further reading
· Python resources
· Generative AI – Resources and additional guidance.
A blended learning journey: live instructor-led training with lab exercises in a top-notch e-learning system. You will keep access to the e-learning system 3-months post-training to revisit the lab exercises and material.
29 labs and 19 case studies.
Standards and references:
OWASP, CWE and Fortify Taxonomy
After attending this training, participants will receive a High Tech Institute certificate.
Receive our monthly updates
