INCIBE Director Warns DeepSeek Could Enable Malware and Cyberattacks
Spain's cybersecurity director warned that the open-source Chinese AI model DeepSeek, due to its low cost and accessibility, could enable widespread creation of malware, disinformation campaigns, and cyberattacks on critical infrastructure. He predicted an increase in AI-related security incidents and called for international cooperation and investment in defensive capabilities.
What happened
Félix Juárez, director of Spain's National Cybersecurity Institute (INCIBE), described DeepSeek as 'a weapon of mass destruction in the field of cybersecurity' during his speech at the CyberTech forum in Tel Aviv on 25 March 2025.
Juárez warned that DeepSeek's low cost and open-source nature make advanced AI capabilities accessible to anyone, including those with minimal resources, enabling them to use the technology for harmful purposes.
Reported impact
Juárez stated that the accessibility of DeepSeek could drive a wave of new threats, including automated creation of sophisticated malware, development of applications to destabilize political systems, and proliferation of fake news and legitimate-looking bots.
He predicted an increase in incidents related to generative AI in the coming year.
Index analysis
DeepSeek, founded in 2021 by former Baidu and Google employees, offers capabilities similar to GPT-4 at zero cost but without the ethical restrictions implemented by Western companies, according to the reports.
The technology removes both technical and economic barriers for malicious actors to generate sophisticated malware campaigns, create convincing disinformation, or develop applications targeting critical infrastructure.
Governance controls
Juárez proposed transforming this threat into an opportunity through international cooperation between public and private sectors, investment in supercomputing and data centres, and accelerated training of cybersecurity specialists.
He advocated for initiatives like NATO's DIANA programme to address the risks.
Classification
Sources and evidence
Sources not yet published
This record was researched and written by the Index from original reporting. Its sources are still being verified by an editor, and they are listed here once that check is complete. Until then the record carries no reviewer.