Few espionage techniques have proven to be as formidable and enduring as the ‘honey trap.’ Across centuries, changing technologies, and evolving geopolitical rivalries, intelligence services have repeatedly relied on personal relationships, emotional manipulation, and sexual leverage to gain access to information that could not be obtained through conventional means. Often dismissed as a cinematic trope populated by glamorous spies and dramatic betrayals, the honey trap remains one of the most enduring instruments in the intelligence toolkit.
History is replete with examples. The biblical story of Samson and Delilah is perhaps the most famous illustration of personal intimacy being used to extract strategic secrets.
Modern popular culture only amplified these narratives, from the femme fatales of early European spy thrillers to the succession of seductresses who populate Cold War fiction and James Bond films. Yet the persistence of such stories is not merely a reflection of artistic imagination. They endure because they draw from methods that intelligence agencies have repeatedly employed in the real world.
Intelligence organisations that openly acknowledge cyber operations, satellite surveillance, and covert funding networks often remain conspicuously reluctant to discuss one of espionage's oldest methods. Despite public denials, virtually every major intelligence service has at some point been accused of using romantic, sexual, or emotional manipulation as an operational tool.
In 2009, Britain's MI5 circulated a 14-page briefing titled ‘The Threat from Chinese Espionage’ warning of sustained Chinese efforts to cultivate Western business leaders and officials through long-term personal relationships that could later be exploited for intelligence purposes. It cautioned that Chinese intelligence operatives had sought to identify and leverage individual vulnerabilities, including sexual relationships, to exert pressure and secure cooperation.
Against this backdrop, recent allegations involving Nvidia CEO Jensen Huang have generated significant international attention. Reports alleging a relationship between Huang and a woman suspected of links to Chinese intelligence have thrust the world's most important AI company into an extraordinary controversy. At the centre of the allegations are broader questions about China's access to advanced semiconductors, US export controls, and the strategic competition surrounding artificial intelligence.
Huang has been among the most prominent voices arguing that Washington should reconsider restrictions on the export of advanced AI chips to China. His argument has been principally economic and technological: denying China access to American chips could encourage Beijing to accelerate domestic semiconductor development and ultimately weaken America's competitive advantage. Under ordinary circumstances, such lobbying would be viewed as part of a broader business debate. However, allegations involving possible intelligence links inevitably cast that advocacy in a different light.
Whether the accusations ultimately prove substantiated or not, the geopolitical implications are considerable. The United States and China are engaged in an increasingly intense contest over artificial intelligence, computing power, and semiconductor supply chains. Nvidia's advanced processors sit at the centre of that competition. Control over cutting-edge AI chips increasingly translates into military, economic, and technological advantages. Consequently, any suggestion that foreign intelligence services may have attempted to influence individuals connected to this ecosystem attracts immediate scrutiny from policymakers and security agencies.
Beyond Nvidia, the allegations spotlight the convergence of human vulnerability, intelligence tradecraft, and the global race for AI dominance.
Honey trapping has also evolved alongside technology. Social media platforms, encrypted messaging applications, video calls, professional networking sites, and increasingly sophisticated AI-generated personas have dramatically expanded the scope of influence operations. Emotional manipulation can now be conducted remotely, continuously, and at scale.
India's recent experiences provide a striking illustration of this transformation. In 2024, Satendra Siwal, an employee at the Indian Embassy in Moscow, was arrested following allegations that he had been honey-trapped online by an individual operating under the identity "Pooja Mehra." Investigators alleged that sensitive military-related information was exchanged after trust had been established through digital communication and financial inducements.
Engineer Nishant Agrawal's conviction in the BrahMos espionage case underscored the dangers posed by social media-based deception. Individuals posing as women on digital platforms allegedly succeeded in extracting sensitive defence-related information, illustrating how online personas can become potent intelligence tools.
Yet sexual leverage is by no means the exclusive province of women. Last year, Haryana-based YouTuber Jyoti Malhotra was arrested over allegations of sharing sensitive information with Pakistani intelligence handlers. Investigators alleged that she had been cultivated by a Pakistani operative and drawn into a relationship that facilitated intelligence collection.
Perhaps the most high-profile example is the case of Pradeep Kurulkar, a senior scientist and former director at DRDO's Research and Development Establishment in Pune. Kurulkar allegedly became entangled in an online relationship with a foreign operative posing as a woman named "Zara" on encrypted messaging platforms. What began as a seemingly personal interaction evolved into a sophisticated intelligence operation, with the operative cultivating trust and emotional dependence before seeking access to sensitive information. Kurulkar shared details of several strategic defence programmes, including missile systems and other critical military projects. The communications took place entirely through digital channels, highlighting how contemporary espionage increasingly relies on virtual engagement rather than physical meetings or traditional spycraft.
The case serves as a stark reminder that in the digital age, the weakest link in any security architecture is often not technology but trust. That reality carries important lessons for governments, corporations, and research institutions. Security can no longer be defined solely by physical barriers or cybersecurity systems. Counterintelligence must place far greater emphasis on human vulnerabilities, behavioural awareness, and digital resilience. Personnel working in sensitive sectors require regular training to recognise manipulation tactics, online grooming efforts, and influence operations. Institutions must also strengthen monitoring mechanisms while remaining attentive to the psychological dimensions of espionage.
From the legends of antiquity to the AI-powered rivalries of the twenty-first century, intelligence services continue to exploit the oldest vulnerability of all, the human desire for connection and intimacy. In an age defined by algorithms and advanced computing, the most valuable target may still be the human being behind the screen.
The author is a security and economic affairs analyst.
The opinions expressed in this article are those of the author and do not purport to reflect the opinions or views of THE WEEK.
