Trend Micro AI Security 2025 warns of a growing threat landscape where thousands of AI servers are misconfigured, unprotected and directly exposed to the internet. A new report from Trend Micro shows that the rapid expansion of AI infrastructure is happening without adequate security practices, opening the door to data theft, manipulation, extortion and other cyberattacks.

AI – opportunity and risk

AI is often touted as the opportunity of the century for businesses globally. With the right security strategy, the technology can revolutionize everything from customer service to research. But if organizations rush into implementation without following best practices, they risk building systems that are more vulnerable than helpful.

– AI may be the opportunity of the century, but without the right security measures, the consequences will be serious. Our report shows that more and more AI systems are being built with insecure and unpatched components – making them extremely vulnerable to threat actors, says Martin Fribrock, Sweden Manager at Trend Micro.

Advertisement

Critical vulnerabilities in AI infrastructure

The report highlights several serious security challenges:

• Vulnerabilities in key components: Exploits have been found in, among other things ChromaDB, Redis, NVIDIA Triton and NVIDIA Container Toolkit, which can be exploited to attack AI models.
• Accidental exposure: Many systems are implemented quickly and without thorough security checks. Trend Micro identified over 200 ChromaDB servers, 2,000 Redis servers, and more than 10,000 Ollama servers which have been openly available on the internet – without authentication.
• Open source: Most AI platforms are open source. This enables rapid innovation, but also means that vulnerabilities can be carried into production systems and become very difficult to detect.
• Container-based weaknesses: AI infrastructure is often containerized, making it vulnerable to the same types of threats that affect cloud and container environments, such as lack of input validation and insufficient execution monitoring.

Thousands of AI servers open the door to cybercriminals

The report warns that these misconfigurations provide cybercriminals with a unique opportunity to exploit flaws in AI systems. Attackers can steal data, manipulate AI models, or use servers as a springboard for larger attacksFor businesses, the consequences can be devastating, including data loss, lost customer trust, and financial losses.

– Many companies underestimate the risksWhen AI servers are exposed in this way, it gives threat actors a direct path into the organization's most sensitive assets, says Fribrock.

Consequences for companies and regulations

For organizations, lack of AI security is not only a technical problem but also a regulatory one. A misconfigured server that leads to data leaks can result in significant fines according to GDPR. In the same way, DORA Regulation within the EU an increasingly important factor, with AI infrastructure now covered by the same digital operational requirements resilience like other critical systems.

Ignoring these requirements can be costly both financially and in terms of brand. Companies that lose control of their AI systems risk harm their reputation, lose customers and have difficulty rebuilding trust.

Recommended actions from Trend Micro

The report proposes a a series of concrete measures to strengthen AI security:

• Regular vulnerability scanning and patching of all AI-related components.
• Continuous inventory of software components, including third-party and open source solutions.
• Safe best practices for container handling, with logging, segmentation and behavioral monitoring.
• Strict configuration controls to ensure that AI servers are not exposed to the internet without authentication.

Balance between innovation and security

The report Trend Micro AI Security 2025 emphasizes that innovation must not come at the expense of security. Companies must find a balance between fast “time to market” and robust cyber protection. Organizations that build security into the foundation of AI systems will both reduce the risk of attacks and strengthen its long-term trust with customers and partners.

– It is crucial that companies find the right balance between innovation and security. Otherwise, AI becomes an attack surface rather than a competitive advantage, summarizes Fribrock.

Conclusion

Trend Micro AI Security 2025 clearly shows that thousands of AI servers are already misconfigured and vulnerable. To avoid data breaches, manipulation of AI models and large-scale cyberattacks Clear security procedures, regular checks and a culture where security is integrated into the entire AI lifecycle are required.

For more information, read the report Trend State of AI Security Report, 1H 2025 here.