Technology news and discussion on TalkFuse.
India flags Chinese-linked AI tools in hacking campaign against financial firms
India’s cybersecurity agency has named a group it calls Crafty Falcon as responsible for a wave of AI-assisted attacks on financial firms. It says publicly available tools helped automate intrusions at “machine speed”, and describes the threat as active and ongoing, amid a rise in attacks on Indian banks and lenders.
CERT-In says the group breached non-bank lenders, fintech companies and cryptocurrency exchanges, but has not identified the victims, the perpetrators or any losses, or linked specific incidents to the campaign. It named four tools, with researchers linking at least two to China; the agency did not say the attacks were state-sponsored. Recent reported incidents include about US$2 million taken through more than 700 unauthorised transfers from two lenders’ accounts, and a leak of 3,897 State Bank of India customer records.
In short
- India says Crafty Falcon used AI tools to attack financial firms.
- CERT-In named four tools, at least two linked by researchers to China.
- Recent Indian financial cyber incidents include US$2 million in unauthorised transfers.
Steelman: both sides, in good faith
The strongest fair case each way — we don't pick a winner.
The case for
If CERT-In’s account is accurate, naming the campaign and warning that accessible AI tools can speed up intrusions may help financial firms strengthen defences before further damage occurs. Public attribution can also draw attention to a broader rise in attacks, even when victims and losses have not yet been disclosed.
The case against
Attribution warrants caution when the agency has not identified the perpetrators or victims, linked specific incidents to the campaign, or established losses, and the tools are publicly available. Reporting Chinese links to some tools without evidence of state sponsorship risks encouraging readers to infer a state connection the agency has not made; precise, qualified communication can protect trust while the evidence is incomplete.
AI Asia Cybersecurity Elections Politics Technology World