NEW DELHI, Nov 5: The Indian Institute of Technology (IIT) Delhi has introduced a six-month online certificate programme in cybersecurity and artificial intelligence (AI) security, aimed at equipping working professionals with the technical knowledge required to tackle emerging digital threats and protect AI-powered systems.
The programme has been launched through the institute’s Continuing Education Programme (CEP) at a time when organisations across industries are increasingly integrating artificial intelligence into their operations, creating new challenges related to data protection, software vulnerabilities, cyberattacks and digital infrastructure security.
The initiative is intended to address the growing demand for professionals who can understand the security risks associated with artificial intelligence and develop effective mechanisms to prevent the misuse of emerging technologies.
According to the programme details, the course will include 78 hours of live online instruction divided into five modules, followed by a capstone project. Classes will be conducted on Saturdays and delivered by IIT Delhi faculty members and industry professionals.
The curriculum has been designed to combine theoretical understanding with practical exposure to cybersecurity challenges involving modern AI systems. Participants will learn how vulnerabilities can emerge in artificial intelligence applications, how attackers attempt to exploit them and what measures organisations can adopt to strengthen their security arrangements.
One of the major areas covered in the programme will be AI red teaming, a security-testing approach in which specialists deliberately examine AI systems to identify weaknesses that could be exploited by malicious users.
Red teaming involves testing how an AI application responds to potentially harmful instructions, attempts to bypass safety restrictions or requests designed to expose sensitive information. The objective is to identify security gaps before the system is deployed or made available to a wider audience.
As businesses increasingly use AI-powered tools for customer support, financial operations, data analysis and internal decision-making, ensuring that these applications function securely has become an important requirement.
The course will also introduce participants to adversarial machine learning, a field that examines how artificial intelligence systems can be manipulated through carefully designed inputs.
Such attacks may cause machine learning models to produce inaccurate results, misclassify information or behave in unexpected ways. These vulnerabilities can create serious challenges for organisations that depend on automated systems for important operational decisions.
Another important subject included in the programme is tool-calling security, which focuses on protecting AI systems that interact with external applications, databases and digital services.
Modern AI assistants can perform tasks by accessing external tools, retrieving information and communicating with different software platforms. However, these capabilities can introduce security risks if the systems are not properly configured.
Unauthorised access, manipulated instructions and poorly controlled permissions could potentially expose sensitive information or allow an AI application to perform actions beyond its intended purpose.
The programme will examine these risks and introduce methods for securing interactions between AI models and external tools.
Participants will also study memory poisoning, a security threat in which attackers attempt to manipulate information stored or retrieved by an AI system. If successful, such attacks could influence future responses or cause a model to rely on misleading information.
The curriculum will further address model supply chain risks, which involve vulnerabilities introduced through the development, distribution and integration of AI models and their supporting components.
AI applications often depend on multiple software libraries, datasets, external services and third-party models. Weaknesses in any of these elements can potentially affect the security and reliability of the complete application.
By covering these subjects, IIT Delhi aims to provide participants with a broader understanding of the challenges involved in developing and maintaining secure artificial intelligence systems.
The course will also introduce several cybersecurity tools and frameworks, including PyRIT, Garak, Sigstore, Open Policy Agent (OPA) and Falco.
These technologies are used for different security-related purposes. PyRIT and Garak help security professionals examine AI systems for vulnerabilities and potential misuse. Sigstore supports software supply chain security by helping verify the integrity and authenticity of software components.
Open Policy Agent provides a framework for implementing policy-based controls, while Falco helps monitor system activity and identify suspicious behaviour.
Familiarity with such tools is expected to help participants understand how security testing and monitoring can be incorporated into practical technology environments.
An important component of the programme will be its capstone project, through which participants will be required to develop and deploy an AI security solution.
The project will allow learners to apply the concepts covered during the course to a practical problem. Their work will be evaluated by IIT Delhi faculty members, providing an opportunity to demonstrate their understanding of AI security and their ability to develop suitable technical solutions.
The programme is particularly relevant for cybersecurity professionals, artificial intelligence and machine learning engineers, data scientists, DevOps specialists, platform engineers and professionals working in governance, risk and compliance.
These professionals are increasingly required to understand not only how digital systems function but also how their security can be maintained as organisations adopt more advanced technologies.
IIT Delhi has specified that applicants should possess a bachelor’s or master’s degree in engineering, computer science, artificial intelligence, information technology, mathematics, statistics or an equivalent discipline.
Candidates must have secured at least 50 per cent marks in their qualifying degree. Familiarity with a programming language will be considered an additional advantage.
Participants who successfully complete the programme will receive a digitally verified e-certificate issued by IIT Delhi’s Continuing Education Programme.
Subodh Vishnu Sharma, Associate Professor in the Department of Computer Science and Engineering at IIT Delhi, said the programme would follow a threat-first approach, helping professionals understand how AI systems are attacked and how appropriate defensive measures can be developed.
He also highlighted the importance of addressing emerging regulatory requirements in India, with the programme providing opportunities to work on practical AI security challenges under the guidance of academic experts and industry practitioners.
The Times of India
The introduction of the programme comes amid a wider transformation in India’s digital technology sector. Artificial intelligence is increasingly being adopted across banking, healthcare, education, manufacturing, retail and government services.
While these technologies offer opportunities to improve efficiency, automate routine activities and support decision-making, their growing use also creates new areas of exposure.
Organisations must ensure that AI applications do not unintentionally disclose confidential information, provide unauthorised access to internal systems or become vulnerable to manipulated instructions.
The growing use of generative AI has also changed the nature of cybersecurity threats. Attackers can potentially use automated technologies to develop convincing phishing messages, analyse publicly available information about organisations and identify weaknesses in digital infrastructure.
At the same time, businesses are increasingly relying on AI systems to process sensitive customer information and interact with internal databases. This makes the protection of AI models, training data and connected software infrastructure an important part of overall cybersecurity planning.
Another challenge involves the security of third-party AI services. Companies that integrate externally developed models and software components must ensure that these technologies meet appropriate security standards and do not introduce weaknesses into their existing systems.
The issue is particularly important for financial institutions and technology companies, where unauthorised access or data manipulation can have serious consequences.
The need for specialised expertise has consequently become an important consideration for organisations planning to expand their use of artificial intelligence.
Traditional cybersecurity training may not always cover the specific vulnerabilities associated with machine learning models, generative AI applications and autonomous digital agents. Professionals increasingly need to understand how these systems process information, interact with external tools and respond to potentially malicious inputs.
IIT Delhi’s initiative seeks to address this requirement by bringing together conventional cybersecurity principles and emerging AI-specific security practices within a structured learning programme.
The inclusion of live instruction, technical tools and a practical project is intended to help participants move beyond theoretical knowledge and develop an understanding of how security measures can be implemented in real-world environments.
The programme also reflects the growing importance of collaboration between academic institutions and the technology industry in developing specialised skills.
As artificial intelligence continues to influence the way organisations operate, educational institutions are being encouraged to update their training programmes to reflect changing industry requirements.
Cybersecurity is expected to remain an important part of this transition, particularly as companies introduce AI-powered applications into business processes that involve sensitive data and critical infrastructure.
For professionals already working in technology-related fields, specialised training in AI security could provide an opportunity to expand their technical capabilities and understand emerging areas of risk management.
The six-month programme therefore represents an effort to prepare professionals for a changing digital environment in which securing artificial intelligence systems is becoming as important as developing and deploying them.
With its combination of technical instruction, industry exposure and practical assessment, IIT Delhi aims to strengthen the availability of specialised expertise needed to address the security challenges accompanying the country’s expanding AI ecosystem.