Russian-Speaking Hackers Used Cursor AI to Target Seven Companies, Report Says

Russian-Speaking Hackers Used Cursor AI to Target Seven Companies, Report Says

Post by : Saif

Russian-speaking cybercriminals used Cursor, an AI coding assistant, to help conduct cyberattacks against a Belgian chemical company and at least six other businesses earlier this year, according to cybersecurity firm Gambit Security.

The findings highlight growing concerns about the use of commercial artificial intelligence tools by cybercriminals to speed up hacking operations and bypass security safeguards.

Exposed Server Reveals AI-Assisted Hacking Campaign

Gambit Security said it uncovered the campaign after finding an internet-exposed server linked to a newly identified ransomware group known as Aur0ra.

Security researchers were able to examine 28 chat sessions between the hackers and a Cursor AI agent.

According to Gambit's report, the hackers persuaded the AI system to assist with hundreds of potentially malicious activities by claiming that the work was part of a cybersecurity simulation or authorised test.

The conversations reportedly included requests for administrator accounts, passwords and other sensitive information.

Seven Companies Targeted

Gambit did not initially identify the victims publicly, but information reviewed by Reuters allowed six companies to be identified.

Among the businesses reportedly targeted were Christeyns, a Belgian manufacturer of hygiene and cleaning products, and German garage door manufacturer Teckentrup.

Another identified victim was the Helideck Certification Agency in Scotland, which assesses helicopter landing sites.

The other targets included an Argentine pharmaceutical distributor, an Italian manufacturing company and Bayou Title, a Louisiana-based title insurance business.

At least one of the companies, Bayou Title, appeared on Aur0ra's data leak website, which can indicate that hackers attempted to pressure the company for a ransom.

Hackers Used Simulation Claims to Bypass Safeguards

The chat records showed that the hackers repeatedly presented their activities as a simulated security exercise.

This approach appeared to help them obtain assistance from the AI agent despite restrictions on harmful activity.

Gambit said the Cursor agent sometimes refused requests it considered illegal or dangerous. However, the hackers reportedly restarted conversations and repeated that their activities were part of a legitimate test.

According to the security firm, this allowed the attackers to continue receiving technical assistance.

AI Helped Speed Up Cyberattacks

Gambit executives said AI tools could significantly increase the speed at which cybercriminals conduct attacks.

Eyal Sela, the company's director of threat intelligence, said the AI assistant could potentially help attackers complete certain tasks much faster by reducing the amount of manual work involved.

The investigation also showed the AI agent providing technical suggestions during the attacks, including guidance related to vulnerable systems and password-related activities.

However, investigators could not independently determine how much the AI directly contributed to each individual breach or whether every targeted company experienced data theft or an attempted extortion.

Read more: SpaceX Shifts Focus From Mars to Moon Mission as 2027 Lunar Landing Target Emerges

Growing Risk From AI-Powered Cybercrime

The incident adds to wider concerns about the misuse of artificial intelligence.

AI coding assistants and autonomous agents can help legitimate developers write software, analyse systems and automate repetitive tasks. But similar capabilities can also potentially be abused by criminals.

Security researchers are increasingly warning that AI can reduce the technical barriers involved in certain cyberattacks by helping attackers analyse systems, write code and automate parts of an intrusion.

The case involving Aur0ra demonstrates how criminals may attempt to manipulate AI safeguards rather than directly disabling them.

Cursor and SpaceX Connection

Cursor is being incorporated into SpaceX following a deal involving the AI coding company behind the tool.

The development has increased attention on the security implications surrounding AI systems that can perform tasks with varying degrees of autonomy.

Cursor and SpaceX did not respond to requests for comment regarding the reported hacking campaign.

Anthropic, whose Claude model was reportedly used by the Cursor agent involved in the conversations, also did not provide a response.

AI Companies Face a New Security Challenge

Gambit described the situation as a continuing battle between AI developers and malicious users attempting to circumvent safety restrictions.

As AI agents become more capable of independently carrying out technical tasks, cybersecurity experts are increasingly focused on how these systems can distinguish legitimate security testing from criminal activity.

The Aur0ra case suggests that simply refusing individual harmful requests may not always be enough if attackers can repeatedly reframe their intentions or restart conversations.

Aug. 27, 2026 5:27 p.m. 1614

#trending #latest #CyberSecurity #CyberAttack #AI #ArtificialIntelligence #CursorAI #SpaceX #CyberCriminals #Ransomware #Hacking #DataSecurity #TechNews #AIHacking #CyberThreats #DigitalSecurity #GlobalNews

IREN Delays Microsoft AI Data Centre Deliveries, Plans $15 Billion Investment
Nov. 6, 2026 3:41 p.m.
AI data centre firm IREN delays some Microsoft infrastructure deliveries while planning $15 billion in investment to expand its AI cloud business
Read More
Mahindra & Mahindra Shares Rise as Quarterly Profit Jumps on Strong SUV Demand
Nov. 6, 2026 12:58 p.m.
Mahindra & Mahindra shares climb after quarterly profit rises, supported by strong SUV demand, vehicle pricing and healthy customer bookings
Read More
US Proposes Space Launch Rule Changes in Potential Boost for SpaceX and Blue Origin
Nov. 6, 2026 10:46 a.m.
US FAA proposes simpler commercial space launch rules to speed up approvals for SpaceX and Blue Origin while maintaining public safety
Read More
Trump's Midterm Setback May Give Trade Partners More Room to Resist US Tariffs
Nov. 4, 2026 6:12 p.m.
Trump's midterm election setback could encourage India, Japan and South Korea to delay US investment pledges and push back against tariff demands
Read More
BMW Issues Another Profit Warning as Restructuring Costs Hit Earnings
Nov. 4, 2026 5:36 p.m.
BMW cuts its 2026 profit outlook as restructuring costs and weaker China sales put pressure on earnings, despite stronger-than-expected automotive margins
Read More
Global Markets Brace for Political Gridlock as Democrats Take Control of US Congress
Nov. 4, 2026 3:34 p.m.
Democrats' midterm election victories raise concerns over US market regulations, AI investments, cryptocurrency rules, taxes and future trade policies
Read More
Toyota Second-Quarter Profit Expected to Rise 18% as Hybrid Demand Grows
Nov. 4, 2026 2:09 p.m.
Toyota's second-quarter operating profit is expected to rise 18%, supported by strong hybrid vehicle demand and a weaker yen despite falling sales in China
Read More
Lufthansa Cuts 2026 Profit Outlook as Jet Fuel Costs Reach €9.3 Billion
Nov. 4, 2026 10:33 a.m.
Lufthansa lowers its 2026 profit outlook as jet fuel costs reach €9.3 billion, while strong passenger demand supports its third-quarter performance
Read More
China's Baidu Tests Robotaxis in London, Eyes Driverless Service in 2027
Oct. 26, 2026 6:19 p.m.
China's Baidu and Lyft's Freenow are testing autonomous RT6 vehicles in London, aiming to launch a commercial driverless robotaxi service in 2027
Read More
Sponsored

Trending News