AI Governance and Responsible AI Policy

BroaeD provides AI consulting, marketing, and revenue infrastructure services for modern B2B companies, including cold email campaigns, outbound systems, inbound workflows, CRM automation, customer support systems, lead qualification, enrichment workflows, reporting, revenue operations automation, campaign strategy, positioning, and AI implementation advisory.

Because these systems may interact with business data, prospect data, customer communications, and third-party platforms, we operate under a responsible AI governance approach designed to protect clients, their customers, and their brand reputation.

Our goal is simple: use AI to reduce operational complexity, improve execution, and scale revenue and marketing workflows without removing human accountability.

Our Responsible AI Principles

1. Human accountability

AI assists our work, but people remain accountable for strategy, advice, approvals, system design, marketing decisions, and final client outcomes. We use human review for important outputs, client-facing materials, campaign logic, automation decisions, strategic recommendations, and workflows that may affect customer experience or brand reputation.

2. Privacy and data protection

We handle client data with care and only use it for the purpose of scoping, building, managing, or improving the agreed client engagement. We do not sell client personal data.

Unless separately agreed in writing, BroaeD does not use client confidential information or client customer data to train public AI models.

3. Transparency

We aim to be clear with clients about when AI, automation, enrichment tools, analytics platforms, or third-party systems are used in a project. Where an AI system is customer-facing or makes automated recommendations, we help define an appropriate disclosure and approval process.

4. Lawful and ethical data use

BroaeD expects that data used in client systems is collected, provided, and processed lawfully. This includes CRM data, lead lists, enrichment data, website visitor data, support tickets, call notes, emails, and other business records.

We design systems and campaigns with attention to applicable privacy, anti-spam, platform, advertising, and data protection requirements.

5. Accuracy and quality control

AI-generated outputs can be incomplete, inaccurate, outdated, or inappropriate. We use review, testing, monitoring, and feedback loops to improve quality and reduce avoidable errors.

6. Brand and reputation protection

We do not treat automation as a reason to lower standards. Outreach, support, sales follow-up, and CRM workflows should protect the client’s domain reputation, customer trust, and market perception.

7. Appropriate automation

We use automation where it improves speed, consistency, and operational quality. We avoid automation that creates unnecessary legal, security, ethical, or reputational risk.

How BroaeD Uses AI

BroaeD uses AI under the direction of qualified team members with relevant experience in AI implementation, automation, revenue operations, marketing, CRM systems, or customer support workflows. AI tools are used to assist expert judgment, not replace it.

AI use is supervised based on the risk and impact of the workflow. Client-facing outputs, strategic recommendations, cold email campaigns, marketing claims, customer communications, CRM automations, support workflows, and other material deliverables are reviewed by a human where required before launch, publication, or client use.

Depending on the engagement, BroaeD may use AI and automation for:

• Ideal customer profile research and segmentation.
• Lead sourcing, enrichment, qualification, and scoring.
• Cold email strategy, sequence drafting, personalization, testing, and campaign operations.
• Personalized outreach drafts and campaign copy.
• Marketing strategy, campaign ideation, creative direction, and content drafting.
• Website, landing page, email, ad, and social media copy assistance.
• Reply classification, follow-up drafting, and routing.
• CRM updates, task creation, pipeline alerts, and reporting.
• Customer support triage, knowledge base assistance, and repetitive ticket handling.
• Customer onboarding workflows and lifecycle automations.
• Sales call summaries, follow-up recommendations, and operational insights.
• Internal productivity, documentation, QA, and workflow design.
• Research and advisory support for AI implementation, tool selection, and workflow design.

When BroaeD refers to systems being “trained on” a client’s ICP, messaging, sales process, or data, this generally means the workflow is configured, prompted, scoped, tuned, connected, or documented around the client’s business context. It does not mean client data is used to train public AI models unless expressly agreed in writing.

What We Avoid

BroaeD does not intentionally use AI systems to:

• Make final legal, financial, medical, employment, insurance, credit, housing, or similarly high-impact decisions.
• Provide legal, financial, medical, tax, or regulated professional advice as a substitute for a qualified professional.
• Send high-risk or sensitive communications without appropriate human or client approval.
• Publish factual, comparative, performance, pricing, or compliance claims without review and client approval.
• Misrepresent a human identity or deceive people into believing they are speaking to a specific person when they are not.
• Use client confidential information to train public AI models without written permission.
• Process sensitive personal data unless it is necessary, lawful, approved, and covered by appropriate safeguards.
• Scrape, access, or use private, paywalled, restricted, or unauthorized data sources.
• Ignore unsubscribe, suppression, consent, or opt-out requirements.

Consulting and Marketing Standards

BroaeD may provide consulting, advisory, strategy, creative, and marketing support. AI may be used to assist with research, analysis, ideation, campaign planning, positioning, copywriting, content drafts, sales enablement, and operational recommendations.

AI-assisted consulting and marketing outputs should be treated as decision support, not final professional advice. Important recommendations, claims, creative assets, campaign messages, and client-facing materials should be reviewed by BroaeD and approved by the client before launch or publication.

For marketing work, BroaeD aims to avoid:

• False, misleading, exaggerated, or unverifiable claims.
• Unauthorized use of third-party intellectual property, confidential material, likenesses, or brand assets.
• Content that impersonates individuals or organizations without permission.
• Dark patterns, deceptive urgency, or manipulative claims.
• Undisclosed synthetic media where disclosure is legally required or important for trust.
• Targeting or personalization that uses sensitive personal data without a lawful basis and appropriate approval.

Clients remain responsible for confirming that their offers, product claims, pricing, guarantees, testimonials, case studies, regulated statements, and industry-specific claims are accurate and lawful.

Data Handling

For client engagements, BroaeD may process information such as business contact data, CRM records, prospect and customer communications, sales activity data, support tickets, website form data, analytics data, and workflow metadata.

We apply practical safeguards, including:

• Limiting access to people and systems that need the data.
• Using approved platforms and vendors for AI, automation, CRM, enrichment, analytics, scheduling, and communication workflows.
• Applying client-specific instructions and access restrictions where agreed.
• Avoiding unnecessary use of sensitive personal data.
• Using secure access methods and reasonable operational controls.
• Supporting deletion, export, or retention requests according to applicable agreements and legal obligations.

Third-Party Tools and AI Vendors

BroaeD may use third-party tools and platforms to deliver services, including AI providers, automation platforms, CRM systems, enrichment tools, email tools, analytics platforms, scheduling tools, and project management systems.

Examples may include OpenAI, Anthropic, Zapier, monday.com, Salesforce, and other approved tools depending on the project scope.

Third-party tools are governed by their own terms, privacy policies, security practices, and data processing commitments. Where required, BroaeD and the client may agree on additional vendor restrictions or a data processing agreement.

Human Review and Approval

BroaeD designs AI systems with review points appropriate to the risk of the workflow.

Examples:

• Draft outreach copy may require human approval before launch.
• AI-generated replies may be approved by a human before sending, or auto-sent only for low-risk, pre-approved response categories.
• CRM updates and lead scores may be logged and reviewed.
• Support automation may include escalation rules for uncertain, sensitive, or high-value cases.
• New automations may be tested before launch and monitored after deployment.

The level of review depends on the client, use case, data sensitivity, workflow risk, and written agreement.

Outbound and Prospecting Standards

For outbound systems and cold email campaigns, BroaeD aims to protect deliverability, brand trust, and legal compliance. This includes attention to:

• Appropriate targeting and segmentation.
• Relevant personalization rather than misleading mass messaging.
• Suppression lists, unsubscribes, and opt-outs.
• Email deliverability configuration, including SPF, DKIM, and DMARC where applicable.
• Avoiding unauthorized or inappropriate data sources.
• Client approval of messaging, positioning, and campaign rules.

Clients remain responsible for confirming that their offers, claims, lists, instructions, and campaign approvals comply with the laws and rules that apply to their business.

Cold Email Campaign Standards

BroaeD may design, build, manage, or advise on cold email campaigns for B2B clients. Because cold email involves prospect contact data, deliverability risk, anti-spam rules, and brand reputation, we treat it as a governed workflow.

For cold email work, BroaeD aims to follow these standards:

• Use relevant B2B targeting rather than indiscriminate mass emailing.
• Use accurate sender identity and avoid deceptive impersonation.
• Avoid false, misleading, exaggerated, or unsupported claims.
• Include a practical opt-out or unsubscribe process where required or appropriate.
• Maintain suppression lists for unsubscribed, bounced, blocked, or disqualified contacts.
• Respect client-provided do-not-contact lists.
• Avoid sending to personal, sensitive, or inappropriate audiences unless there is a lawful basis and client approval.
• Use approved data sources and avoid unauthorized, private, or restricted data.
• Use deliverability safeguards such as SPF, DKIM, DMARC, mailbox setup, domain reputation monitoring, bounce monitoring, and volume controls where applicable.
• Review campaign copy, targeting, personalization logic, sending domains, and escalation rules before launch.
• Monitor for bounce rates, spam complaints, negative replies, deliverability issues, and unusual sending behavior.

AI may be used to support cold email research, segmentation, personalization, copy drafting, reply classification, and follow-up suggestions. AI-generated outreach should be reviewed before launch, and fully automated sending should only be used within approved campaign rules.

Clients remain responsible for approving the final target audience, offer, claims, messaging, list sources, sender identity, campaign rules, and legal/compliance position for the markets where emails are sent. This may include laws such as CAN-SPAM, GDPR, PECR, CASL, or other local anti-spam and privacy rules depending on the geography and audience.

Marketing Content and Campaign Standards

For marketing strategy, content, advertising, landing pages, sales enablement, or campaign work, BroaeD may use AI to support ideation, drafting, research, editing, personalization, and analysis.

Before publication or launch, clients should review and approve:

• Brand positioning and messaging.
• Product, service, pricing, ROI, performance, and guarantee claims.
• Case studies, testimonials, logos, and customer references.
• Regulated or industry-specific statements.
• Paid advertising copy, targeting, and landing page claims.
• Public-facing AI-generated or AI-assisted content.

BroaeD does not guarantee any specific marketing result, including traffic, rankings, impressions, leads, conversion rates, sales, revenue, or return on ad spend, unless expressly stated in a written agreement.

Customer Support and AI Agent Standards

For customer support, onboarding, and AI agent workflows, BroaeD aims to design systems that improve response speed while preserving customer trust.

Where appropriate, support systems should include:

• Clear escalation paths to a human.
• Guardrails for refunds, account issues, legal questions, security matters, medical or financial claims, and other sensitive cases.
• Approved knowledge sources.
• Testing before launch.
• Monitoring after deployment.
• Client ownership of final policies, support rules, and customer commitments.

Risk-Based Governance

BroaeD classifies AI and automation use cases by risk.

Low-risk examples:

• Internal note summarization.
• Drafting non-sensitive documentation.
• Organizing tasks.
• Basic reporting assistance.

Medium-risk examples:

• Prospect personalization.
• CRM enrichment.
• Lead scoring.
• Support ticket triage.
• Sales follow-up drafts.

Higher-risk examples:

• Customer-facing AI agents.
• Automated emails sent without human review.
• Workflows using sensitive personal data.
• Systems that affect access to important services, pricing, eligibility, employment, finance, healthcare, housing, insurance, or legal rights.

Higher-risk workflows require additional scoping, approvals, testing, documentation, and monitoring. BroaeD may decline or require changes to work that creates unacceptable legal, security, ethical, or reputational risk.

Monitoring and Incident Response

BroaeD may monitor AI and automation systems for errors, routing issues, hallucinations, inappropriate outputs, data problems, failed automations, or unexpected behavior.

If a material issue occurs, BroaeD will take reasonable steps to:

• Investigate the issue.
• Reduce or pause the affected workflow where appropriate.
• Notify the client when the issue materially affects the engagement.
• Correct configuration, workflow, prompt, access, or process failures.
• Support any legally required notifications according to the applicable agreement.

Client Responsibilities

Clients are responsible for:

• Providing accurate instructions, business context, and approval criteria.
• Ensuring they have the right to use any data, lists, content, credentials, or materials they provide.
• Reviewing and approving important AI-assisted outputs before use, unless a different approval process is agreed in writing.
• Complying with privacy, anti-spam, industry, platform, and sector-specific laws that apply to their business.
• Maintaining appropriate access to their own systems, domains, email accounts, CRM, and third-party tools.
• Not providing sensitive or restricted data unless it is necessary, lawful, approved, and covered by appropriate safeguards.

Framework Alignment

BroaeD’s AI governance approach is informed by recognized responsible AI and AI risk management practices, including principles found in NIST AI RMF and ISO/IEC 42001. This does not mean BroaeD is certified under those frameworks unless expressly stated

Contact

Questions about BroaeD’s AI governance approach can be sent to:

Email: legal@broaed.com

Website: https://broaed.com/contact