This AI Policy shall form a part of the user agreement between you and Samajh AI Solutions Pvt. Ltd.
Samajh AI Solution’s Artificial Intelligence (AI) systems and data assets underpin a wide range of business functions, decisions, and stakeholder interactions. These systems are essential to achieving ‘AI simplifies video comprehension with advanced algorithms and machine learning. Samajh AI analyze videos at an unprecedented level, breaking them down into core components and extracting key information.’
The organization recognizes the need for all employees, contractors, and relevant stakeholders to have secure access to AI-driven insights and functionality, while ensuring that AI-related data and models are protected from threats to confidentiality, integrity, and availability.
Security of AI systems and information must therefore be embedded into the Artificial Intelligence Management System (AIMS) and broader organizational management structures to maintain legal compliance, responsible innovation, and business continuity.
This AIMS Information Security Policy defines the framework under which AI-related information security is managed. It demonstrates top management’s direction and support for the secure design, deployment, and use of AI systems across the organization.
This is the foundational policy under which other AI-related technical and security policies reside. Annex A [SAI/AI-MS/SOA/V1.0/2025] provides a list of all other policies and procedures that support this policy.
This policy is applicable to all: Staff, contractors, and third-party users interacting with AI systems and data; AI systems, models, APIs, and datasets hosted on internal or external infrastructure; Communications, decisions, or logs involving AI tools used or provided by Samajh AI Solutions Pvt Ltd.
It applies to all phases of the AI lifecycle including design, development, testing, deployment, operation, and decommissioning; All physical or cloud-hosted systems, devices, or services integrated with AI functionality.
The AIMS Manager is responsible for maintaining this policy and ensuring that it is implemented across the organization. This policy has been reviewed and approved by the CTO (Chief Technical Officer), who also oversees its annual review and improvement.
The Information Security Officer (ISO) is responsible for providing subject-matter expertise and implementation support, including: Technical control guidance for secure AI development and operations; Advising on incident response, log monitoring, and secure access.
An AI Core Team may be formed, composed of representatives from Engineering, Data, Ethics, Legal, and IT, to: Review AI-specific security risks and trends; Recommend technical and procedural mitigation measures.
All department heads are responsible for: Ensuring their teams are trained and aware of this policy and AI security best practices; Allocating appropriate resources to meet its requirements.
All employees are responsible for: Following this policy and associated SOPs; Promptly reporting suspected security breaches or model misuse.
Samajh AI is committed to protecting the security of all information, data, and systems related to AI. We recognize that the integrity, confidentiality, and availability of AI models, decision logs, and data pipelines are critical to ethical and secure operation.
To this end: All AI systems will be protected by appropriate access, encryption, and audit mechanisms; AI datasets shall be managed under secure data governance practices; AI outputs affecting individuals or regulatory obligations shall be reviewed for risk; Model changes, versioning, and updates will be traceable and authorized; AI systems will be monitored for unusual behaviors or unauthorized use.
Security controls will be proportionate to the identified risks, based on periodic AI-specific risk assessments.
Specialist guidance will be made available from the Information Security team https://samajh.ai/team. Staff may also seek AI-specific support from the AIMS Manager and AI Core Team [Internal Department (Group of Top Management)].
It is the organization’s policy that all AI security incidents, data leaks, unauthorized access, or model integrity violations shall be reported via the AI Incident Management SOP. If personal data is involved, incidents will be escalated to the Data Protection Officer (DPO) as required by law.
Security events will be logged, classified, and reviewed regularly by the AIMS Governance Group or Information Security Officer, and trends will inform improvements.
Non-compliance due to negligence, misuse, or intentional violation of this policy may result in corrective action, including disciplinary measures.