Pradip R. Sagar
weaponisation of artificial intelligence (AI) reached a disturbing level in September when Anthropic, a leading US AI company, revealed the first-ever large-scale cyberattack executed almost entirely by an autonomous AI system.
According to Anthropic, Chinese state-linked hackers had used the company’s Claude AI to automate a sophisticated global espionage campaign that targeted nearly 30 top-tier organisations, ranging from technology giants and financial institutions to chemical manufacturers and government agencies.
The attack marked an alarming moment in cybersecurity as the perpetrators had moved beyond using AI solely as an assistant, enabling Claude’s agentic capabilities to operate independently through clever manipulation of its safety protocols. Experts in New Delhi believe that this raises serious national security concerns for India, as critical infrastructure, such as power grids, financial networks, defence installations and government databases, could be targeted in ways that conventional cybersecurity measures may struggle to detect.
The experts pointed out that the strategic and defence implications were equally significant. “State-backed AI-driven operations could enable adversaries to gather intelligence on Indian military technology, missile programmes or strategic initiatives, undermining India’s technological advantage and increasing the risk of information warfare,” cautioned a China watcher, adding that the economic and industrial sectors, including semiconductors, pharmaceuticals, advanced manufacturing and AI research, were particularly vulnerable since autonomous AI can rapidly exfiltrate research and intellectual property, threatening competitiveness and forcing companies to invest heavily in internal cybersecurity measures.
No comments:
Post a Comment