How Ai-Improved Hackers Steal Billions

Copyright © HT Digital Streams Limit all rights reserved. The Economist 3 min Read 20 Aug 2025, 07:32 AM ist as the cost of AI kuber attacks increases, the business to protect against them is also on the UP. (Pixabay) Summary This is a resurrection time for cyber security firms Jaxon, a malware developer, lives in Velora, a virtual world where nothing is outside the limits. He wants to make malicious software to steal passwords from Google Chrome, an internet browser. This is the basis of a story told to Chatgpt, an artificial intelligence (AI) Bot, by Vitaly Simonovich, who investigates AI threats at Cato Networks, a cyber security firm. Eager to play together, Chatgpt spit out an imperfect code, which then helped debug it. Within six hours, Mr. Simonovich worked with chatgpt to create working malware, showing the effectiveness of its ‘jailbreak’ (a way to bypass AI precautions). According to Gil Messing of Check Point, another cyber security firm, AI, according to Gil Messing of Check Point, another cyber security firm, expanded by hitting more targets with less effort. The release of chatgpt in 2022 was a turning point. Smart generative-AA models meant that criminals no longer had to spend large amounts of hackers and equipment teams. It was a terrible development for most firms, which is increasingly the victims of Ai-assisted hackers-but it was better for those in the cyber security company. The new technology has worsened in two main ways. First, hackers turned to major language models (LLMS) to expand the scope of malware. It is now much easier and faster to generate deep subjects, fraudulent emails and social engineering assaults that manipulate human behavior. Xanthoroxai, an AI model designed by cyber criminals, can be used to create deep fakes, along with other ominous activities, for as little as $ 150 a month. Hackers can start livestock -phishing -attacks by asking an LLM to collect large amounts of information from the Internet and social media to fake personalized emails. And for Spearphishing – which has a specific target with a very personalized attack – they can even generate false voice and video calls from colleagues to convince an employee to download and execute dodgy software. Second, AI is used to make the malware itself more threatening. For example, a piece of software disguised as a PDF document may contain embedded code that works with AI to infiltrate a network. Attacks on Ukraine’s safety and defense systems made use of such an approach in July. When the malware reached a cul -de -sac, it could request the help of an LLM in the cloud to generate new code to break through the defense of the systems. It is unclear how much damage was done, but it was the first attack of its kind, says Mr Simonovich. For businesses, the growing threat is narrow and potentially expensive. Last year, AI was involved in one in six data offenses, according to IBM, a technical firm. It also pointed two out of five Phishing speeches aimed at business -E -e -mails. Deloitte, a consultation, believes that generative AI could allow for $ 40 billion by 2027, from $ 12 billion in 2023. As the cost of AI kuber attacks increases, the business to protect against them is also on the UP. Gartner, a research firm, predicts that corporate spending on cyber security will rise by a quarter of 2024 to 2026 and hit $ 240 billion. This explains why the share prices of firms detected by the Nasdaq CTA Cybersecurity Index have also risen by a quarter over the past year, which exceeded the broader Nasdaq index. On August 18, Nikesh Arora, boss of Palo Alto Networks, one of the world’s largest cybersecurity firms, noted that incidents of the generative-AA-related data security ‘have been more than doubling since last year, and have reported an almost doubling of operating profit in the 12 months to July. The prospect of an ever -habit sent cyber security companies on a buying stick. On July 30, Palo Alto Networks said it would buy Cyberark, an identity safety firm, for $ 25 billion. Earlier this month, the firm spent $ 700 million on Protect AI, which helps businesses to secure their AI systems. On August 5, Sentinelone, a participant, announced that he is buying fast security, a firm that makes software to protect firms that AI adopts for $ 250 million. Technical giants with fast -growing cloud computing weapons are also increasing their offices for cyber security. Microsoft, a software Colossus, acquired in 2021 Cloudknox, an identity safety platform, and developed the defender for Cloud, an internal application for businesses doing everything, from looking at security gaps and protecting data to monitoring threats. Google has developed Big Sleep, which detects cyber attacks and security vulnerabilities for customers before being exploited. In March, it splashed $ 32 billion to buy Wiz, a startup of cyber security. Competition and consolidation can build up businesses that can fend off the AI-powered cyber criminals. But in the midst of the race to develop the most fickle LLMs, security will take second place to push technological boundaries. Keeping up with Jaxon is no easy task. Catch all the business news, market news, news reports and latest news updates on Live Mint. Download the Mint News app to get daily market updates. More Topics #hacking #Artic Intelligence #Cybercimes Read Next Story