Safety controls are security controls that are designed to protect human life and physical assets from harm or damage. Examples of safety controls include fire alarms, sprinklers, emergency exits, backup generators, and surge protectors. Safety controls should fail open, which means that they should remain operational or allow access when a failure or error occurs. Failing open can prevent or minimize the impact of a disaster, such as a fire, flood, earthquake, or power outage, on human life and physical assets. For example, if a fire alarm fails, it should still trigger the sprinklers and unlock the emergency exits, rather than remain silent and locked. Failing open can also ensure that essential services, such as healthcare, transportation, or communication, are available during a crisis. Remote access points, logging controls, and logical security controls are other types of security controls, but they should not fail open in a data center. Remote access points are security controls that allow users or systems to access a network or a system from a remote location, such as a VPN, a web portal, or a wireless access point. Remote access points should fail closed, which means that they should deny access when a failure or error occurs. Failing closed can prevent unauthorized or malicious access to the data center's network or systems, such as by hackers, malware, or rogue devices. Logging controls are security controls that record and monitor the activities and events that occur on a network or a system, such as user actions, system errors, security incidents, or performance metrics. Logging controls should also fail closed, which means that they should stop or suspend the activities or events when a failure or error occurs. Failing closed can prevent data loss, corruption, or tampering, as well as ensure compliance with regulations and standards. Logical security controls are security controls that use software or code to protect data and systems from unauthorized or malicious access, modification, or destruction, such as encryption, authentication, authorization, or firewall. Logical security controls should also fail closed, which means that they should block or restrict access when a failure or error occurs. Failing closed can prevent data breaches, cyberattacks, or logical flaws, as well as ensure confidentiality, integrity, and availability of data and systems. Reference: CompTIA Security+ Study Guide: Exam SY0-701, 9th Edition, page 142-143, 372-373, 376-377

Smishing is a type of phishing attack that uses text messages or common messaging apps to trick victims into clicking on malicious links or providing personal information. The scenario in the question describes a smishing attack that uses pretexting, which is a form of social engineering that involves impersonating someone else to gain trust or access. The unknown number claims to be the company's CEO and asks the employee to purchase gift cards, which is a common scam tactic. Vishing is a similar type of attack that uses phone calls or voicemails, while phishing is a broader term that covers any email-based attack. Reference: CompTIA Security+ Study Guide: Exam SY0-701, 9th Edition, page 771; Smishing vs. Phishing: Understanding the Differences2

This scenario describes a bug bounty program, which invites external security researchers to responsibly identify and report vulnerabilities in an organization's systems. CompTIA Security+ SY0-701 defines bug bounty programs as structured initiatives that reward researchers for discovering and disclosing security flaws before they can be exploited by malicious actors.
Bug bounty programs extend security testing beyond internal teams and provide continuous, real-world testing of public-facing assets. They are particularly effective for identifying zero-day vulnerabilities, logic flaws, and edge-case issues that automated tools may miss.
Red teaming (B) is a controlled, internal or contracted adversarial exercise, not an open invitation. Open- source intelligence (C) involves gathering publicly available information, not vulnerability reporting. Third- party information sharing (D) refers to sharing threat intelligence, not soliciting vulnerability reports.
Because the CISO is inviting the broader security community to report vulnerabilities, the correct answer is A: Bug bounty.

The correct answer is B. Air-gapped.
An air-gapped architecture physically or logically isolates systems from other networks, including the public internet and internal local networks. Because the devices cannot directly connect to the internet or local network, updates must be transferred manually, often using removable or portable media.
This aligns with CompTIA Security+ SY0-701 security architecture concepts involving segmentation, isolation, high-security environments, and protection of sensitive or critical systems.
Why the other options are incorrect:
A . Microservices
Microservices is an application architecture where software is broken into smaller independent services. It does not describe network isolation.
C . Software-defined networking
Software-defined networking separates the control plane from the data plane and allows centralized network management. It does not inherently require devices to be disconnected from the internet or local network.
D . Serverless
Serverless is a cloud computing model where the cloud provider manages the underlying infrastructure. It does not describe isolated firmware update architecture.
Therefore, the best answer is air-gapped.

A jump server is a device or virtual machine that acts as an intermediary between a user's workstation and a remote network segment. A jump server can be used to securely access servers or devices that are not directly reachable from the user's workstation, such as database servers. A jump server can also provide audit logs and access control for the remote connections. A jump server is also known as a jump box or a jump host12.
RADIUS is a protocol for authentication, authorization, and accounting of network access. RADIUS is not a device or a method to access remote servers, but rather a way to verify the identity and permissions of users or devices that request network access34.
HSM is an acronym for Hardware Security Module, which is a physical device that provides secure storage and generation of cryptographic keys. HSMs are used to protect sensitive data and applications, such as digital signatures, encryption, and authentication. HSMs are not used to access remote servers, but rather to enhance the security of the data and applications that reside on them5 .
A load balancer is a device or software that distributes network traffic across multiple servers or devices, based on criteria such as availability, performance, or capacity. A load balancer can improve the scalability, reliability, and efficiency of network services, such as web servers, application servers, or database servers. A load balancer is not used to access remote servers, but rather to optimize the delivery of the services that run on them . Reference = How to access a remote server using a jump host Jump server RADIUS Remote Authentication Dial-In User Service (RADIUS) Hardware Security Module (HSM)
[What is an HSM?]
[Load balancing (computing)]
[What is Load Balancing?]