Lexis+ AI vs Westlaw AI Quiz

Answers and explanations appear as you take the quiz.

1. What technique do Lexis+ AI and Westlaw's AI research features both use to ground their answers?
   • Answering only from what the language model memorized during training
   • Retrieval-augmented generation that searches the vendor's own legal collection before the model writes an answer
   • Copying answers from public web search results
   • Having human editors write each answer in real time
2. In legal research AI, what is misgrounding?
   • Citing a real case for a point it does not actually support
   • Citing a case that does not exist
   • Failing to return any answer
   • Returning results from the wrong database vendor
3. What did the 2024 Stanford study find about Lexis+ AI and Westlaw's AI-Assisted Research?
   • Neither tool ever produced an incorrect answer
   • Only Lexis+ AI produced errors
   • Both gave incorrect or misgrounded answers on a meaningful share of questions, with Westlaw's tool erring more often in that test
   • The tools performed identically on every question
4. Which editorial assets does the guide associate with Westlaw rather than Lexis?
   • Shepard's and Lexis Practical Guidance
   • Protégé and Shepard's
   • Lexis headnotes and Protégé
   • The West Key Number System and KeyCite
5. Why does the guide recommend including false-premise questions in your test set?
   • Because they make the test shorter
   • Because a model may accept the premise and answer as if it were true
   • Because false-premise questions are required by bar rules
   • Because they only work on one of the two platforms
6. If the retrieval stage misses the controlling case, what does the guide say can happen?
   • The system automatically refuses to answer
   • The citator flags the missing case in red
   • The answer can be fluent and wrong
   • The model searches the open web instead
7. Which company did Thomson Reuters acquire in 2023, bringing the CoCounsel assistant into its products?
   • Casetext
   • Lex Machina
   • Clarivate
   • Bloomberg Law
8. Why do multi-step, agent-style research features make checking more important rather than less?
   • Because agents cannot cite sources
   • Because they only work on federal law
   • Because they remove citator flags from results
   • Because longer outputs contain more claims to verify