From the course: CompTIA SecAI+ (CY0-001) Cert Prep
Unlock this course with a free trial
Join today to access over 25,500 courses taught by industry experts.
MCP servers
From the course: CompTIA SecAI+ (CY0-001) Cert Prep
MCP servers
The Model Context Protocol, or MCP, is a standard, open-source approach that allows AI applications to interact with external tools simply and securely. Organizations that want to expose their services to AI applications build an MCP server and provide resources to those tools. This may include access to data or tools for interacting with other services. For example, Microsoft provides the Azure MCP server that allows AI applications to interact with Azure cloud services and resources. Splunk offers an MCP server that allows AI applications to explore data stored in Splunk, run queries, and leverage Splunk AI services. From a security perspective, MCP servers can help enforce the principle of least privilege for AI workflows. Rather than letting an AI application connect directly to a SIEM, EDR platform, ticketing system, or vulnerability scanner, an MCP server can sit in the middle and offer a small set of well-defined actions. The AI doesn't need full API keys or broad credentials…
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.
Contents
-
-
(Locked)
The AI lifecycle1m 39s
-
(Locked)
Business alignment in the AI lifecycle1m 43s
-
(Locked)
Data collection2m 20s
-
(Locked)
Data preparation3m 15s
-
(Locked)
Model development and selection2m 13s
-
(Locked)
Model evaluation and validation2m 29s
-
(Locked)
Model deployment and integration3m 25s
-
(Locked)
Monitoring and maintenance3m 19s
-
(Locked)
-
-
(Locked)
Manipulating application integrations4m 8s
-
(Locked)
AI supply chain attacks2m 4s
-
(Locked)
Insecure plug-in design2m 9s
-
(Locked)
Insecure output handling1m 23s
-
(Locked)
Output integrity attacks2m 8s
-
(Locked)
Model denial of service1m 31s
-
(Locked)
Excessive agency1m 33s
-
(Locked)
Overreliance1m 34s
-
(Locked)
AI hallucinations1m 4s
-
(Locked)
-
-
(Locked)
Monitoring prompts and responses2m 51s
-
(Locked)
Log monitoring4m 30s
-
(Locked)
Rate and cost monitoring5m 1s
-
(Locked)
Auditing for AI hallucinations3m 33s
-
(Locked)
Auditing for accuracy3m 29s
-
(Locked)
Auditing for bias and fairness4m 35s
-
(Locked)
Auditing access and security compliance3m 48s
-
(Locked)
-
-
(Locked)
Responsible AI5m 29s
-
(Locked)
AI risks2m 23s
-
(Locked)
Introduction of bias2m 37s
-
(Locked)
Accidental data leakage2m 53s
-
(Locked)
Reputational loss2m 11s
-
(Locked)
Accuracy and performance of the model2m 22s
-
(Locked)
Intellectual property risks3m 31s
-
(Locked)
Autonomous systems2m 27s
-
(Locked)
Shadow IT and shadow AI1m 48s
-
(Locked)
Awareness training2m 21s
-
(Locked)