An AI hallucination is when a large language model (LLM) generates false information. Hallucinations largely happen due to three causes: poor data quality, the AI’s training and generation methods, and unclear input context.
It will take about 1-4 minutes to generate the AI summary because we do a lot more discovery and processing compared to a traditional Westlaw search, including sending relevant materials to the language model for analysis.
Yes. Prior conversations will be available on the left-hand side of the search screen during a session and can also be found in a user’s search history.
365 days
The time out period is 6 hours, however some browsers can override with their own inactivity timeouts.
System requirements for the new Westlaw are:
Westlaw and Practical Law products employ a Retrieval Augmented Generation (RAG) approach, grounding the Large Language Model (LLM) in Thomson Reuters' content when generating a response. Responses include source citations, enhancing transparency and allowing users to easily verify the underlying language supporting the response. Thomson Reuters strongly recommends customers validate all responses.
No, getting started is specific to the user.
No. The AI does not summarise cases. CoCounsel does have a summarisation capability.
No. The URL has not changed from Westlaw Australia to Westlaw Precision Australia. Even after switching to Westlaw Precision Australia, all deeplinks will continue to function.