US-Iran war: Anthropic reveals Iranian operators used Claude to gather intel on US naval forces, Israeli targets

As the US-Iran war is set to enter its seventh month, tensions continue to mount across the Middle East. Against this backdrop, Iran-linked operators allegedly used Anthropic’s Claude AI to collect intelligence and prepare targeting recommendations aimed at US naval forces in the region.

Iranian operators use Claude to collect information on US naval assets

According to a new report released by the artificial intelligence (AI) company on Thursday (local time), Anthropic said it had identified an Iran-linked account that used its Claude chatbot to collect and analyse publicly available information on US naval assets. The account also used the AI model to obtain "targeting handbooks" and track naval positions using open-source information.

The company said, “We identified and disrupted an Iran-nexus threat actor that used Claude to collect and analyze publicly accessible data to develop targeting recommendations against US naval forces in the region. The threat actor used Claude to compile targeting handbooks through a Python pipeline the threat actor built with Claude's assistance to identify and track naval positions based on open-source information.”

It added, "The compiled material included a roster of US personnel scraped from captions on public military photographs; publicly accessible ship and aircraft transponder identifiers; commercial satellite-imagery query scripts; and an inventory of public websites that exposed US naval movements. The threat actor also directed Claude to compile vulnerability research on shipboard systems, cataloging known CVEs in maritime VSAT terminals, Cisco communications equipment, and industrial control products."

However, the AI company’s report does not indicate whether the Iran-linked operators successfully used Claude to attack a US Navy vessel or naval base. Instead, it describes how the AI model was used to gather and analyse publicly available information that could potentially help develop detailed targeting intelligence on US naval forces.

Iran used Claude to target Israel: Here's what the report revealed

Anthropic, in its report, also stated that it identified an Iran-nexus threat actor that used the AI chatbot to build an automated, open-source intelligence identity-profiling harness targeting Israeli governmental and non-governmental individuals, and Jewish diaspora organisations. Separately, the actor used Claude to make it harder to tell that its malware was malware.

In one area of activity, the threat actor allegedly used Claude to coordinate an open-source intelligence and reconnaissance tool that profiled hundreds of individuals in Israel and the Jewish diaspora. The actor also ran an automated identity-profiling system to enrich an existing target list and generate open-source intelligence on Israeli and US individuals. Claude was used to speed up the collection and analysis of publicly available data.

Anthropic's findings come as the US-Iran war fuels a wider regional confrontation, with Iran-aligned Houthis stepping up attacks on Saudi Arabia and tensions continuing to flare between Israel and Iran-backed groups across the region.

US-Iran war escalates

The war launched by the US and Israel against Iran in late February has evolved into a deeper conflict, with tensions spreading across the region. Initially expected to last only a few weeks, the war is still far from over, with the prolonged stalemate proving costly for both sides. An agreement reached in June quickly fell apart, and there has been little sign of diplomatic progress since. Low-level fighting continues, while the US appears to have no clear exit strategy, AP reported.