AI Use & Responsible AI Policy
This policy explains how Chaitanya Digital Solutions may use artificial intelligence and AI-assisted technologies in its services, internal operations, research, automation and digital systems, together with the principles and limitations that guide responsible use.
Effective Date: 5 September 2026
AI Should Improve the System, Not Remove Accountability
CDS uses AI as a technology capability within defined business contexts. AI may assist with information processing, research, content workflows, operational tasks, analysis, automation and decision support. The use of AI does not by itself transfer responsibility for business, legal, security or operational decisions to an AI system.
Useful Over Impressive
AI is used where it can provide a practical improvement to a workflow, information process, system or business operation. AI is not introduced merely for novelty.
Human Accountability
Human review remains important for outputs or actions where accuracy, context, business consequences, security or legal considerations require human judgement.
Defined Context
AI workflows should operate within an identified purpose, relevant information, business rules and appropriate operational boundaries.
Controlled Automation
Automation should be proportionate to the task. Where an automated action could create material consequences, appropriate review, controls or approval mechanisms may be required.
Where CDS May Use AI-Assisted Technology
Depending on the service, project requirements and available technology, AI-assisted systems may be used in several areas. Not every service uses AI, and the specific tools used may vary over time.
Research & Analysis
AI may assist with research organisation, information summarisation, pattern identification, classification and preliminary analysis.
Content Workflows
AI may assist with drafting, restructuring, ideation, summarisation, content analysis, metadata preparation and other content-related workflows.
Business Automation
AI may be integrated into workflows involving information processing, task routing, response assistance, classification or other repetitive operational activities.
Knowledge Assistance
AI-assisted systems may help organise or retrieve information from approved business knowledge sources and support internal operational workflows.
Customer & Response Assistance
Where appropriate, AI may assist with drafting or organising responses. Human intervention may remain necessary depending on the context and consequences of the response.
Decision Support
AI may provide suggestions, classifications, summaries or recommendations to assist a human decision-maker. Such output should not automatically be treated as a final decision.
A Controlled AI Workflow
The appropriate level of human review depends on the purpose, sensitivity, reliability requirements and potential impact of the particular workflow.
AI Processing Does Not Override Privacy Responsibilities
Information used in an AI-assisted workflow remains subject to the applicable privacy, contractual, security and data-handling requirements governing the relevant service.
Purpose Limitation
Information should be used for a defined and relevant business purpose rather than being supplied to an AI system without a reasonable operational need.
Data Minimisation
Where practical, AI workflows should use only the information reasonably necessary for the intended task.
Client Data
Where CDS processes client information on behalf of a client, the applicable service agreement, privacy requirements and agreed data-handling arrangements remain relevant.
Third-Party AI Platforms
Some AI capabilities may depend on third-party providers, APIs, software platforms or hosted infrastructure. Their own terms, privacy practices and technical limitations may apply.
Human Judgement Remains Part of the System
AI systems can generate useful outputs while still producing incomplete, incorrect, outdated, ambiguous or contextually inappropriate information. Human review should therefore be applied where the consequences of an incorrect output justify it.
Review Before Material Action
Outputs that may materially affect a client’s business, customers, security, finances, legal position or public communications may require human review before action.
Context Verification
AI-generated information should be checked against relevant source material when factual accuracy or context is important.
No Blind Automation
AI-generated suggestions should not automatically be treated as authoritative simply because they were produced by an AI system.
Escalation Where Needed
A workflow may require escalation to a qualified human, client representative, technical specialist, legal professional or other appropriate authority depending on the issue.
AI Output Is Not Automatically a Fact
AI systems operate probabilistically and may produce outputs that contain errors, omissions, unsupported conclusions or information that requires additional verification.
Accuracy
CDS does not represent that AI-generated output will always be accurate, complete or suitable for a particular purpose.
Current Information
AI output may not automatically reflect the latest information unless the workflow includes an appropriate current-data source or verification process.
Bias & Context
AI systems may reflect limitations or biases present in their underlying models, training information, inputs or surrounding workflow.
Professional Decisions
AI-assisted output should not replace appropriate professional advice or qualified judgement where a matter requires it.
Where Additional Care Is Required
AI-assisted workflows involving sensitive information, security, financial activity, legal matters, health-related information, identity, employment, access control or other consequential decisions may require additional controls and human oversight.
Clients Also Have a Role in Responsible AI Use
Where an AI-assisted system is implemented for a client, responsible operation may depend on the information, instructions, permissions and review processes supplied by the client.
Provide Appropriate Information
Clients should provide accurate and relevant information and identify restrictions that may apply to particular data or workflows.
Identify Sensitive Context
Clients should inform CDS when a proposed workflow involves information requiring special handling or additional controls.
Review Important Outputs
Where the agreed workflow requires human approval, clients remain responsible for reviewing and approving relevant outputs before consequential action.
Use Systems Lawfully
Clients remain responsible for using implemented AI systems, data and outputs in accordance with applicable law, contractual obligations and their own organisational policies.
AI Security Is Part of Digital Security
AI-assisted systems can introduce additional considerations involving credentials, prompts, connected applications, APIs, documents, integrations and generated outputs.
Access Control
Access to AI-enabled systems and connected resources should be limited according to the requirements of the workflow.
Credentials & Secrets
API keys, passwords, access tokens and other confidential credentials should not be unnecessarily exposed to AI systems or included in ordinary prompts.
Connected Systems
Integrations between AI systems and external applications should be configured with appropriate permissions and controls.
Monitoring & Review
Where appropriate, AI workflows may benefit from logging, review, testing, access controls and periodic reassessment.
AI Should Not Be Used to Circumvent Responsibility
AI capabilities provided through CDS should not be intentionally used for unlawful, abusive, fraudulent, deceptive, malicious or unauthorised activity.
Do not use AI systems to facilitate unauthorised access, credential theft or malicious activity.
Do not intentionally use AI workflows for fraud, impersonation, deception or unlawful manipulation.
Do not use AI systems to process information without the required authority, permission or lawful basis.
Do not treat AI output as a mechanism for avoiding applicable professional, legal, contractual or organisational obligations.
AI Providers and External Platforms
CDS may use third-party AI models, APIs, software, hosting infrastructure or other technology providers when appropriate for a service or workflow.
Provider Terms
Third-party AI providers may have separate terms, acceptable use requirements, privacy policies, retention practices and technical limitations.
Technology Changes
AI models, capabilities, pricing, interfaces and provider policies may change over time. A workflow may therefore require periodic review or modification.
Availability
CDS does not guarantee uninterrupted availability of any third-party AI platform or external technology dependency.
Alternative Architecture
Where practical, a workflow may be designed so that a change in a particular AI provider does not automatically eliminate the broader business process.
AI Policy Works With the Rest of the CDS Legal Framework
This policy should be read together with the other applicable CDS policies and the specific commercial terms governing a client engagement.
Questions About AI Use?
Contact CDS if you need clarification about an AI-assisted workflow, data-handling requirement, human review process or technology dependency associated with a service.
Chaitanya Digital Solutions
Mumbai, Maharashtra, India
Phone: 02269718012
WhatsApp: +91 7249595642
Email: admin@chaitanyadigitalsolutions.com
Website: chaitanyadigitalsolutions.com
Responsible AI Starts With Clear Boundaries.
CDS uses AI as part of a broader digital infrastructure approach, with practical controls, appropriate human oversight and respect for privacy, security and applicable obligations.
Last Updated: 5 September 2026