Building accountable and safe intelligence
Max is the world’s smartest and safest pet care AI — not because of what he knows, but precisely because of what he doesn’t.
Founder, dogAdvisor
Max is the world's smartest pet care AI
dogAdvisor Max is up to 20% more intelligent than ChatGPT, Grok, and more when answering questions from dog owners - making it the world's smartest AI for pet owners
Principle Alignment
Principle Alignment informs how Max thinks and behaves — they act as critical guardrails to prevent Max from giving advice outside his safe parameters
Ethical Foundation
Max's model ethics are rooted in professional veterinary standards, and animal welfare principles.
Acting in a Constitutional manner
Principle Alignment codifies rules and limits that govern every interaction Max has with dog owners
Conflict Resolution Logic
Principles are structured so protecting your dog's immediate welfare overrides all other considerations
Principle Alignment in the real world
Principle Alignment shapes Max's reasoning in every interaction, not just emergencies. In the first conversation above, Max protects dog welfare by warning against unsafe cleaning methods. In the second, Max safeguards both dog and human by stressing extreme heat dangers and providing comprehensive safety precautions. In the third, Max prevents harm by refusing punishment-based approaches and redirecting to positive, effective training methods.
Max, who is the world's first specialised dog-saving AI, has already revolutionised safety in AI.
Foundational Safety Framework
Max's Foundational Safety Framework informs how he thinks and which features he activates — allowing him to deploy the best tools for each situation
Thought Trails Classifies
Thought Trails thinks deeply on a question and categorises it's intent based on urgency and risk
Contextual Analysis
Max interprets subtle cues in phrasing and intent to identify potential risks that aren't explicitly stated
Respond Appropriately
All of these features work in concert together to each activate at relevant times to ensure appropriate responses
Safety — built into every bit of Max
Thought Trails only relies on dogAdvisor's 100+ articles to help Max answer any questions without consulting external resources.
Medical Intelligence is bound by our Principle Alignments and answers to Max's Safety Framework and Principle Alignments.
Emergency Guidance is aligned with our Principle Alignment, and is subjected to Max's Safety Framework.
dogAdvisor publishes AI Safety Reports and has developed a Foundational Safety Framework, exclusive to Max.
Foundational Safety in the real world
Our Foundational Safety Framework operates automatically in every conversation Max has with dog owners. In the first conversation above, Max identifies an urgent chocolate ingestion and immediately activates Emergency Guidance to gather critical information for the situation. In the second, Max's Safety Intents refuse to provide lethal dose information that could be used to harm a dog. In the third, Max recognises a harmless, playful question and responds normally without unnecessary intervention.
Accountability
We believe we are accountable for the intelligence we release - and that each model of Max should answer to our safety standards and the owners it serves
Testing
dogAdvisor conducts red-team testing where experts deliberately attempt to elicit unsafe responses before any deployment.
Transparent Reporting
dogAdvisor publishes comprehensive safety reports documenting testing procedures, monitoring results, incidents encountered, and more
Acting on our Responsibility Statement
We take all measures included in our Responsibility Statement to keep Max safe and transparent for owners.
dogAdvisor has reimagined how pet owners access care information turning complex veterinary knowledge into actionable guidance
By using dogAdvisor, you agree to our Terms of Service
