Responsible AI

ClaroCam uses artificial intelligence to analyze images and video frames to provide information, identification, and guidance. Our AI can identify objects, translate text, assess safety hazards, provide repair guidance, analyze multi-angle video captures, and much more.

• ClaroCam is an informational tool — not a replacement for qualified professionals
• Our AI provides assessments and suggestions, not diagnoses or certifications
• All safety, medical, and veterinary outputs include disclaimers to consult professionals
• We do not make autonomous decisions — humans always decide what to do with the information
• Video scan mode captures multiple frames for more comprehensive analysis — all frames are discarded after processing

When it comes to safety, health, and security features, our AI is built to err on the side of caution. We believe a false warning is always better than a false "all clear."

• The AI will never say "all clear" unless it is highly confident — when in doubt, it warns
• Medical features always recommend seeing a doctor or pharmacist for verification
• Safety scans report everything questionable — we let professionals make the final call
• Return Home Safe will flag anything suspicious rather than risk a missed threat
• Pill identification always includes "verify with your pharmacist" guidance
• We would rather give you a false alarm than miss something that could hurt you

AI is powerful but imperfect. We are transparent about what our technology can and cannot do.

• AI analysis is based on visual information only — it cannot see what is hidden, smell, or touch
• Confidence levels are provided with every result (high, medium, low)
• When uncertain, the AI presents multiple possibilities rather than guessing
• Image quality, lighting, and angle affect accuracy — the AI will tell you if it needs a better photo
• Our AI knowledge has a training cutoff — very recent products or events may not be recognized
• We continuously improve accuracy but no AI system is 100% correct 100% of the time

Your photos and data are treated with respect. We believe in minimal data collection and maximum privacy.

• Photos are sent to the AI for analysis and are not permanently stored on our servers
• Your images are not used to train AI models
• Scan history is stored in your personal account — only you can see it
• Workforce team scans are visible only to your team admin and members
• Near-miss reports are intentionally anonymous — no user attribution
• We do not sell your data to third parties
• You can delete your account and all associated data at any time

ClaroCam is designed for positive, productive use. We have safeguards against misuse in both individual scans and the community feed.

• The Sketch Pad will not process drawings depicting violence, hate symbols, or explicit content
• The AI will not provide instructions for creating weapons or harmful substances
• Inappropriate requests receive a polite refusal with suggested positive alternatives
• We do not acknowledge or describe harmful content — we simply decline and redirect
• Community feed posts are subject to moderation — harmful or misleading content may be removed
• These safeguards protect our community without restricting legitimate creative or professional use

Every person deserves the same level of protection and thoroughness from our safety tools.

• Safety assessments are equally thorough regardless of who is using the app
• Workforce tools provide the same standard of inspection for every company size
• Our AI does not make assumptions about users based on personal characteristics
• Hazard detection is objective — based on regulations and physics, not judgment of people
• Accessibility features (Read Aloud, voice controls) ensure the app works for everyone

AI should inform and assist — never replace human judgment in critical decisions.

• ClaroCam provides information — you decide what to do with it
• Safety features recommend calling professionals (911, doctors, mechanics, electricians)
• Workforce reports are designed for review by qualified safety managers
• We never encourage users to skip professional consultation based on AI results
• The Emergency Response feature helps you share information with first responders — it does not replace calling 911

If you encounter inaccurate, harmful, or concerning AI behavior, we want to know.

• Email us at privacy@clarocam.net to report AI accuracy issues
• Report any instance where safety advice seemed incorrect or dangerous
• Let us know if content moderation missed something it should have caught
• Your feedback directly improves the system for everyone
• We take every report seriously and investigate promptly