security researcher, former malware analyst
Joined June 2012
- Tweets 268
- Following 146
- Followers 97
- Likes 827
12 Photos and videos
May 28
42.6% of MCP server repos with confirmed exploitable vulnerabilities show signs of AI-generated code.
With Alfredo Oliveira at TrendAI Research, we swept 19,000 MCP servers; analysis shows 600–1,650 repos contain exploitable vulnerabilities.
trendmicro.com/vinfo/us/secu…
21
David Fiser retweeted
Apr 28
MCP adoption is exploding as AI agents plug into everything, but security is lagging badly. Exposed MCP servers nearly tripled to 1,467 in just months, many running with zero authentication and leaking direct access to cloud services. Read more: research.trendmicro.com/48qJ…
3
364
David Fiser retweeted
Apr 15
[ZDI-26-261] (0Day) Docker Desktop credentialHelper Directory Traversal Privilege Escalation Vulnerability (CVSS 7.5; Credit: Nitesh Surana (niteshsurana.com) & Nelson William Gamazo Sanchez of Trend Research) zerodayinitiative.com/adviso…
1
2
596
Mar 26
Your AI Gateway Was a Backdoor: Inside the LiteLLM Supply Chain Compromise | Trend Micro (US) trendmicro.com/en_us/researc…
1
1
110
Mar 25
Heads up for #RSAC2026 attendees — at #CloudVillage, Alfredo Oliveira (Sr. Security Researcher, TrendAI) is presenting: "Mitigating MCP Server Threats in Cloud Environments" Worth the stop by 1:50PM Schedule: cloud-village.org/rsa26 #MCPSecurity #CloudSecurity #TrendAI
2
32
David Fiser retweeted
The TrendAI™ State of AI Security Report is out.
As the AI threat landscape evolves, organizations must adopt proactive strategies to address unique risks from supply chain vulnerabilities and exposed infrastructure.
Strengthen your AI defenses: research.trendmicro.com/Stat…
2
3
1,346
23 Sep 2025
Beware of AI applications and their privacy policy claims. In new research, we reveal when security measures fail miserably, resulting in the exposure of sensitive user photos and proprietary AI models due to unsecured cloud storage.
trendmicro.com/en_us/researc…
24
17 Sep 2025
As part of our continuous research on how to strengthen Model Context Protocol (MCP) security we explore how proper MCP server hardening within containers can mitigate security risks and offer security benefits essential for protecting cloud workloads. trendmicro.com/vinfo/us/secu…
23
13 Aug 2025
Don't forget that secret management best practices apply to MCP servers! trendmicro.com/vinfo/us/secu…
1
2
113
17 Jul 2025
Don't be a victim of fast and unsafe MCP server deployments. MCP Security: Network-Exposed Servers Are Backdoors to Your Private Data trendmicro.com/vinfo/ph/secu…
22
28 May 2025
Trend Micro Leading the Fight to Secure AI | Trend Micro (US) trendmicro.com/en_us/researc…
33
David Fiser retweeted
29 Apr 2025
Organizations running AI-powered services with proprietary models could be at risk of intellectual property theft, particularly if their models or inference services are exposed through misconfigured APIs.
Here’s what you need to know: research.trendmicro.com/4jyY…
1
1
410
David Fiser retweeted
17 Apr 2025
The exposure of OpenVPN and SSH keys within a container image can lead to full network compromise. Learn how to protect your keys and certificates today.
Read more here:⬇️ research.trendmicro.com/4iUh…
3
5
479
David Fiser retweeted
13 Apr 2025
Cybercriminals can weaponize stolen certificates and private keys to infiltrate systems undetected. Learn how to safeguard your organization from these risks.
Here’s what you need to know: ⬇️ research.trendmicro.com/4iUh…
1
2
518
David Fiser retweeted
7 Apr 2025
[ZDI-25-205] Amazon AWS CloudFormation Templates Uncontrolled Search Path Element Remote Code Execution Vulnerability (CVSS 9.8; Credit: Nitesh Surana (@_niteshsurana) of Trend Micro Research) zerodayinitiative.com/adviso…
2
9
908
3 Apr 2025
From Registries to Private Networks: Threat Scenarios Putting Organizations in Jeopardy trendmicro.com/vinfo/ph/secu…
15
David Fiser retweeted
13 Mar 2025
[ZDI-25-144|CVE-2025-23243] NVIDIA Riva gRPC API Missing Authentication for Critical Function Authentication Bypass Vulnerability (CVSS:6.5; Credit: David Fiser and Alfredo Oliveira (Nebula of Trend Micro)
zerodayinitiative.com/adviso…
7
11
3,368
David Fiser retweeted
13 Mar 2025
[ZDI-25-145|CVE-2025-23243] NVIDIA Riva Triton Inference Server Missing Authentication Vulnerability (CVSS:7.3; Credit: David Fiser and Alfredo Oliveira (Nebula of Trend Micro)
zerodayinitiative.com/adviso…
1
6
10
3,749
David Fiser retweeted
2 Jan 2025
Exposed container registries allow attackers to upload tampered images, posing a threat to AI systems and daily operations.
Our analysis talks about this and more to help you learn how to secure your AI systems:⬇️ research.trendmicro.com/4gAd…
2
3
619
12 Dec 2024
Do you want to know if there is a ticking bomb in your DevOps environment? Is your AI pipeline safe? Read our newest research paper wrote in cooperation with Alfredo Oliveira.
trendmicro.com/vinfo/us/secu…
33