- The intelligence report for anthropic threats describes the acceleration of AI attacks
- AI now feeds all parts of the cyber attack process
- Such an attack was identified during the “piracy of atmospheres”
One of the largest AI companies in the world, Anthropic, warned that his chatbot had been “armed” by the actors of the threat of “committing a large -scale theft and an extortion of personal data”. The report on the threats of anthropic details the means by which technology is used to carry out sophisticated cyber attacks.
The army AI makes pirates faster, more aggressive and more successful – and the threat report emphasizes that ransomware attacks which would have previously required training years can now be designed with very little technical skills.
These cyber attacks are lucrative for pirates, AI being now used for fraudulent activity such as theft of information on credit cards and identity theft, attackers even using AI to analyze stolen data.
“Ambrance hacking”
Defenders have long warned that AI reduces obstacles to cybercrime, allowing low -skilled hackers to make complex attacks, but LLMs now help criminals at each point along the attack process.
The report describes a particular threat that it survives “hacking”, which refers to a campaign in which Claude was used to scale and build an extortion diagram. The name is a reference to the software development method “coding of the atmosphere which is based heavily on the AI to generate code and create applications.
The Cluade Code Execution Environment has been used; “Automate the recognition, harvesting of diplomas and the penetration of the large -scale network, potentially affecting at least 17 separate organizations during the last month of the government, health care, emergency services and religious institutions.”
Anthropic surveys have revealed that cybercriminals were targeting a range of sectors, focusing on data theft and extortion. These attacks led to “compromise of personal files, in particular health data, financial information, government identification information and other sensitive information, direct ransom requests sometimes exceeding $ 500,000”.
