Testing Anthropic’s Claude Sonnet 4.5: clinical study docs & franchise agreements | Model Behavior

We test the newest AI model, Claude Sonnet 4.5, against real-world enterprise datasets, including clinical study documents and a franchise legal agreement.

Watch how Claude 4.5:

→ Identifies critical inconsistencies across multiple documents with precise citations
→ Provides concise, easy-to-understand summaries
→ Offers detailed legal reasoning for contract revisions
→ Improves significantly over Claude Sonnet 4 in industries like hospitality, life sciences, legal, and retail

Discover how Claude Sonnet 4.5 is unlocking key enterprise workflows by delivering trustworthy, auditable AI insights. If you’re interested in AI advancements in enterprise document analysis, contract review, or legal workflows, this demo is a must-watch.

Timestamps:

0:00 Introduction to Claude 4.5 testing
0:45 Clinical study document analysis
1:35 Inconsistency identification with citations
2:15 Franchise agreement revisions for franchisee benefit
3:00 Summary and comparison with Claude 4.0

See our full evaluation of Anthropic’s Claude 4.5 here: https://blog.box.com/first-look-claud...