Immediate Steps to Prevent AI Scams: Jan Wuppermann, NTT DATA
Here are some immediate steps you can take to protect yourself from AI-generated scams:
Be mindful of the personal information you share online - Some Generative AI systems like may leverage input data for training and result in information leakage. Avoid disclosing personal information on social media or during AI conversations and choose AI systems that has stringent data privacy rules.
Verify AI outputs against trusted sources - AI might provide seemingly convincing but inaccurate outputs. Scammers may use this to produce convincing phishing emails or false news. Users should verify AI-generated data against trustworthy sources and be careful when considering urgent actions or sensitive information requests.
Recognise and Resist Social Engineering Tactics - AI is improving phishing and social engineering techniques. Producing personalised messages that resemble authentic communications is now simpler. Users should be aware of phishing indications such odd email addresses, unwanted demands, or grammatical problems. Check requests via trusted sources before responding.
Educate Yourself on Adversarial Prompts - Cyber criminals can use crafted prompts to manipulate AI into producing biased or malicious outputs. Users should understand how generative AI systems work and avoid relying entirely on AI-generated advice or decisions, particularly in high-stakes scenarios.
Leverage AI-powered security tools for enhanced protection - Generative AI can protects users too. AI-driven security tools can help identify threats and vulnerabilities, decreasing scam risk.
"By adopting these steps, users can mitigate risks associated with generative AI while benefiting from its capabilities." advises Jan Wuppermann, Chief Digital Officer for APAC at NTT DATA. He also advised users to stay vigilant against scams and staying informed about emerging threats and security solutions.
The challenge of identifying AI-generated content is becoming more complex, according to Jan. "These models, such as those based on Generative Adversarial Networks (GANs) and Transformers, are trained on extensive datasets to replicate human-like patterns in text, images, and other media. Their outputs can often appear indistinguishable from authentic human-created content," he explains. Recent insights into generative AI's capabilities and limitations highlight the tendency of models to "hallucinate" or generate factually incorrect but convincing content, increasing the complexity of the problem.
AI Content Detection Strategies
Jan identifies three key strategies to enhance AI content detection: developing specialised machine learning algorithms to identify AI patterns, implementing explainable AI for better transparency and accountability, and embedding standardised metadata within AI outputs to distinguish them from human content—a practice already adopted in various global regulatory frameworks. "At NTT DATA, we recognise the importance of addressing these challenges and advancing generative AI responsibly. Our expertise in natural language processing and intelligent document processing allows us to design solutions that emphasise transparency, reliability, and collaboration between AI and humans. This commitment aligns with our methodology to harness generative AI’s immense potential while mitigating its risks." claimed Jan. He added that the detecting and managing AI-generated content requires a balanced approach, balancing cutting-edge technology with ethical considerations for responsible deployment across industries. Building trust and transparency through strong governance frameworks is a crucial in the rapidly evolving GenAI space.
Changes in Financial Institutions
In 2025, Jan predicts that financial institutions will use AI to deliver "intelligent, anticipatory customer
experiences that go far beyond conventional service levels." They will transform their services through AI-driven personalisation, offering one-to-one financial guidance based on customer interaction data, while expanding into broader ecosystem services including insurance, tax, legal, and lifestyle offerings, and delivering concierge-style experiences that provide preferred access to events and travel planning to enhance customer loyalty. To offer more AI services, financial institutions are also exposing themselves to greater security risks. As such, Jan expects the institutions to also likely implement AI-driven solutions to help protect users against fraud, particularly through advanced biometric authentication systems that analyse voice, facial, and behavioural patterns to identify potential deepfake manipulation. These protective measures work alongside proactive anomaly detection systems that leverage AI's strategic capabilities - proven in gaming and recommendation systems - to automate real-time fraud detection and key decision-making processes.
AI Continues to Grow Amid Challenges
Despite growing concerns about AI scams, inaccuracies, and privacy risks, AI adoption is not expected to slow down significantly, according to Jan "Nearly 99% of businesses surveyed are planning further investments in Generative AI (GenAI), recognising its immense potential to drive improvements in productivity, security, compliance, and more." While acknowledging the challenge of trust, he notes, "We expect companies to shift increasingly effort into this topic simply to protect their own brand as much as to comply with further developing regulatory requirements," adding that explainable AI is crucial to restoring trust. Explainable AI (XAI) is a key focus for NTT DATA as it enhances transparency by making generative models more interpretable and accountable. This means that XAI makes it possible to understand how AI models arrive at their decisions, it helps reduce bias in AI systems by making it possible to identify and correct any biases present in the data or algorithms.