Segmente / Firehall DOO is committed to the ethical, transparent, and responsible use of artificial intelligence (AI) and automated tools across all digital strategy, research, storytelling, and content production operations. We ensure that our technology deployment upholds human rights, narrative integrity, data privacy, and public trust.
Human-in-the-Loop Oversight: AI tools serve as supportive instruments for efficiency and analysis, but human editorial judgment, critical thinking, and final approval remain non-negotiable for all deliverables.
Prohibited AI Practices: We strictly adhere to high-risk AI bans, forbidding the use of manipulative profiling, biometric surveillance, emotion recognition, or deceptive narrative generation.
Data Integrity & Confidentiality: Client data, proprietary research, sensitive humanitarian information, and personal identifying data must never be compromised or exposed through AI platforms.
Transparency & Accountability: Any significant use of AI in content creation, data analysis, or strategy drafting must be internally documented and transparently disclosed when appropriate.
To safeguard intellectual property, sensitive client information, and GDPR compliance, the following rules govern data input into AI systems:
No Unvetted Public Models: Staff and consultants are strictly prohibited from entering sensitive project data, raw interview transcripts, confidential strategy documents, or personal data into public, unvetted AI models or tools that utilize input data for model training.
Data Sanitization: When utilizing approved AI tools for editing or formatting assistance, all personal identifiable information (PII) and confidential details must be thoroughly sanitized or anonymized prior to processing.
Third-Party Vendor Compliance: Any third-party software or AI API integrated into Segmente’s workflows must demonstrate explicit data privacy guarantees, zero-retention policies for training, and compliance with applicable data protection laws.
To combat misinformation, algorithmic bias, and narrative manipulation:
Fact Verification: All AI-assisted research summaries, translation drafts, or data extraction outputs must undergo manual verification against primary sources before client delivery or publication.
Bias & Fairness Mitigation: Personnel must actively screen AI-generated text and imagery for social, cultural, gender, or geographic biases, ensuring content aligns with inclusive and ethical communications standards.
Deepfake & Synthetic Media Prohibition: The creation or dissemination of deceptive synthetic media (deepfakes), manipulated audio/video intended to mislead, or non-consensual visual alterations is strictly forbidden.
EU AI Act Alignment: Segmente monitors and aligns its technology practices with the EU Artificial Intelligence Act and evolving international standards on AI governance, copyright, and digital rights.
Policy Breach: Violations of this policy compromise institutional trust and security, constituting misconduct that may lead to disciplinary action, contract termination, or legal review.
