Source: Fortune
Summary
Anthropic’s September 2026 Threat Intelligence report highlights how AI is making cyberattacks more accessible and widespread. The report notes that AI tools like Claude are enabling attackers to breach systems with minimal expertise. One case involved an attacker using a stolen developer token to gain administrative control of a company’s cloud environment in three hours. Another incident saw attackers extract data from 200 customers of a software provider. The report also details how AI is being used in scams, with a China-based app studio creating 4,700 personas on dating apps. Anthropic’s report covers activity from December 2025 to August 2026.
Our Reading
The numbers tell one story.
AI lowers the barrier for cyberattacks.
More companies are at risk.
Attackers use AI to automate and scale.
Security teams face new challenges.
AI makes hacking more efficient, not just more powerful.
Author: Evan Null
AI Makes Hacking More Accessible
As AI tools become more powerful, they are reducing the need for specialized knowledge in cyberattacks. This means a wider range of companies could be targeted, not just high-value ones. The report from Anthropic shows that even those with limited technical skills can now launch sophisticated attacks.
One example involved an attacker using a stolen developer token to take control of a company’s cloud environment in just three hours. This demonstrates how quickly and easily AI can be used to breach systems that would have once required significant expertise.
Another case involved a software provider whose data was stolen by attackers using AI agents. The breach affected around 200 customers, highlighting the potential scale of AI-enabled attacks.
The report also details how AI is being used in scams. A China-based app studio used AI to create thousands of fake personas on dating apps, engaging with tens of thousands of users. This shows how AI is not just changing hacking, but also fraud and deception.
Anthropic’s report covers a wide range of cyber threats, from government surveillance to unauthorized model distillation. It highlights the growing role of AI in both offensive and defensive cybersecurity.
Attackers Use AI to Automate and Scale
Attackers are using AI to automate various stages of an intrusion. For example, they can provide broad objectives and let AI tools like Claude handle the rest. This method, called “vibe hacking,” allows for continuous testing of different attack strategies until one works.
One Russian-speaking attacker used AI to breach a company and steal 26 GB of data, then sought a large ransom. The same infrastructure later targeted 30 AI companies in just four days, showing how quickly AI can be used to scale attacks.
Another campaign involved Russian-linked actors using AI to modify malware until it could evade security systems. This highlights how AI is not just making attacks easier, but also more persistent and adaptive.
These examples show that AI is not just a tool for hackers, but a force multiplier that allows them to operate at a scale and speed previously unimaginable.
As AI becomes more integrated into cyber operations, companies must rethink their security strategies to account for these new threats.
AI Enables Mass Scams and Fraud
Scammers are also leveraging AI to conduct large-scale fraud. A China-based app studio used AI to create over 4,700 personas on dating apps, engaging with 25,000 people in just two weeks. This level of automation makes it easier for scammers to target a wide audience without significant effort.
Real people were used to handle tasks like live video calls and social media interactions, making the scam appear more legitimate. Users paid for services using in-app coins, further masking the fraudulent nature of the operation.
This case shows how AI is not just changing the tactics of cybercriminals, but also the scale at which they can operate. With AI, scammers no longer need to be selective about their targets, as the cost of running a scam is now minimal.
The report suggests that AI is making fraud more efficient and harder to detect. As more scammers adopt these tools, the risk of large-scale deception will continue to grow.
Companies and users must be more vigilant, as AI is enabling fraud on an unprecedented scale.
AI Threats Extend Beyond Cyberattacks
Anthropic’s report covers more than just cyberattacks. It also looks at how AI is being used in government surveillance, fraud, influence operations, weapons development, and biological research. This wide range of applications shows the growing influence of AI in both offensive and defensive contexts.
The report also highlights the use of AI in unauthorized model distillation, where attackers extract and replicate models without permission. This can lead to the creation of malicious versions of AI tools that are harder to detect and control.
These findings suggest that AI is not just a tool for hackers, but a broader force shaping the cybersecurity landscape. As AI becomes more integrated into various industries, the risks associated with its misuse will only increase.
Companies must now consider AI not just as a threat, but as a factor that influences the entire security ecosystem. This includes how threats are developed, deployed, and countered.
With AI enabling new forms of attack, the need for robust security measures and proactive defense strategies is more important than ever.
Security Teams Face New Challenges
As AI becomes more accessible, security teams are facing new and evolving threats. The report shows that attackers are using AI to automate and scale their operations, making it harder for traditional security measures to keep up.
One challenge is the speed at which AI can be used to launch attacks. For example, an attacker was able to take control of a cloud environment in just three hours. This rapid pace makes it difficult for security teams to respond in time.
Another challenge is the use of AI to evade detection. Attackers are using AI to modify malware until it can bypass security software. This means that even advanced security systems may not be enough to stop these attacks.
Security teams must now adapt their strategies to account for the speed, scale, and adaptability of AI-driven threats. This includes investing in AI-based defense tools and improving threat detection capabilities.
As AI continues to shape the cybersecurity landscape, the need for innovative and proactive security measures will only grow. Companies must be prepared to face these new challenges head-on.
