AI Security projects for B.Tech students — real examples

What students actually build

• Capstone projects that ship AI features are expected to consider guardrails and safe input handling.
• Smart India Hackathon entries increasingly involve AI components where safety is part of the build quality.

Labs and infrastructure

• The AICTE IDEA Lab provides GPU workstations used for AI model and AI-application experiments.
• Lab work runs on institute-provided systems under faculty supervision.

How VSET teaches AI Security

AI security is the practice of protecting machine learning and LLM-based systems from misuse, manipulation and unsafe behaviour — covering guardrails, input handling, model safety and the security of the surrounding application. It is a fast-emerging area at the boundary of AI and security engineering. At VSET this maps to documented coursework depth inside B.Tech CSE (Cyber Security).

• AI-security topics including prompt injection, guardrails and model safety are covered as coursework relevant to production AI systems at VSET.
• This is possible because VSET runs parallel AI tracks — B.Tech CSE (AI & ML) and B.Tech CSE (AI & DS) — alongside the dedicated cyber security track.
• The material connects application security practice to machine learning system design.
• Framing is defensive throughout: designing systems that behave safely under adversarial input.