Responsible AI Principles
The TJ Branding Responsible AI Framework
TJ Branding applies the following principles when designing or operating AI-enabled workflows.
Defined purpose
We identify the business purpose of an AI workflow before deploying it and seek to limit the system to functions appropriate for that purpose.
Human accountability
AI may assist a decision-making process, but responsibility remains with TJ Branding, our client, or another authorized human decision-maker as appropriate. We do not intentionally delegate consequential professional or business decisions to AI where human judgment is required.
Risk-proportionate oversight
Not every AI use case presents the same level of risk. We apply greater review, testing, documentation, and human involvement where an AI system could materially influence customer decisions, client strategy, professional recommendations, public communications, or other higher-impact activities.
Data minimization
We seek to collect, use, and provide AI systems only the information reasonably necessary for the intended purpose. We discourage the unnecessary submission of highly sensitive personal information to TJ Branding's public AI tools.
Transparency
Where appropriate and legally required, we identify when a person is interacting directly with an AI-powered experience or when AI materially contributes to an automated output.
Testing and monitoring
AI-enabled systems may be tested for factors including: functional performance; output relevance; unsupported or misleading claims; edge cases; inappropriate recommendations; privacy risks; security issues; usability; integrations and delivery failures; And adherence to intended scope.
Escalation and intervention
Continuous improvement
AI technologies, laws, standards, and model behavior change rapidly. We periodically reassess our tools, vendors, processes, and controls and may modify our practices as technology and regulatory requirements evolve.