29 March 2025, New Delhi: As Artificial Intelligence (AI) applications continue to reshape the digital landscape, India’s top cyber security agency has issued a stern warning to users about the potential risks associated with these emerging technologies. The Computer Emergency Response Team of India (CERT-In) has cautioned that not all AI apps are safe and has flagged the increasing vulnerabilities associated with AI-powered platforms.
In a recent advisory, CERT-In, which operates under the Ministry of Electronics and Information Technology (MeitY), has highlighted that users signing up for AI applications should exercise caution and consider using anonymous accounts not linked to their personal or professional identities. The agency’s advisory is part of a broader effort to protect users and the Indian cyberspace from rising cyberattacks and malicious exploitation in the digital era.
Mounting AI Security Concerns
The federal agency’s advisory underscores that AI apps, while offering significant advancements and convenience, are susceptible to various forms of cyberattacks. The warning comes amid the exponential rise in the use of AI-driven chatbots, image generators, voice assistants, and productivity tools, which have become integral to the daily lives of millions.
“Not all AI applications out there are safe,” CERT-In’s advisory reads, pointing to growing concerns over security loopholes that can be exploited by cybercriminals and adversarial entities.
The advisory details several technical vulnerabilities inherent in AI systems, including:
- Data Poisoning: The manipulation of training data by attackers to introduce biases or errors.
- Adversarial Attacks: Tactics where attackers input specially crafted data to deceive AI models and manipulate their outputs.
- Model Inversion: Techniques used by malicious actors to reconstruct sensitive information from AI models.
- Prompt Injection: A method of embedding malicious instructions into inputs that can influence AI behavior.
- Hallucination Exploitation: Manipulating AI systems to produce false or misleading information.
These vulnerabilities, CERT-In notes, stem from flaws in AI design, training, and the interaction mechanisms of the applications.
Guidelines for Safer AI Use
In light of these threats, CERT-In has recommended several precautionary measures for users engaging with AI applications:
- Use Anonymous Accounts: Avoid linking AI app accounts to personal, financial, or professional identities.
- Check for Permissions: Be cautious about the amount of personal data being shared with AI platforms.
- Monitor Outputs: Verify information generated by AI apps, especially when used for decision-making or content creation.
- Stay Informed: Keep abreast of known vulnerabilities and security patches related to AI services.
The advisory also appeals to developers and technology providers to strengthen their platforms’ security architecture and adopt responsible AI development practices to mitigate risks.
Growing Relevance Amid AI Boom
This warning by CERT-In gains significance as India rapidly embraces AI across sectors such as healthcare, finance, governance, and education. The adoption of AI technology has grown exponentially over the past few years, but the security risks have largely remained underreported or underestimated by general users.
Cybersecurity experts have welcomed the advisory, stating that public awareness about the dark side of AI is long overdue. “AI applications are evolving at a breakneck pace, but so are the methods of attackers. This is a timely reminder that convenience must not come at the cost of data security and privacy,” said a senior cybersecurity analyst.
As India positions itself as a global leader in AI adoption and innovation, experts emphasize the need for a balanced approach—one that harnesses the potential of AI while building robust defenses against its misuse.
CERT-In’s advisory serves as a timely wake-up call to users and developers alike: technology’s power must be matched by caution, responsibility, and awareness.
Source: Press Trust of India
