Get in touch
Frequently Asked Questions
Here are the most common questions we get about Spellbook.
What dataset is Spellbook trained on?
Spellbook is built on OpenAI's GPT-3 large language model. It has been trained on a massive dataset of 42 terabytes of text from the Internet as a whole, contracts, books and Wikipedia. Spellbook is further tuning the model using proprietary legal datasets.
Is Spellbook aiming to give advice to non-lawyers?
No. Spellbook is strictly available for legal professionals today. It is a tool to help professionals work much more efficiently, very similar to GitHub Copilot for programming!
What if Spellbook gives bad legal advice?
Spellbook's output should always be reviewed by a lawyer. It is not meant to provide advice to end-clients or to replace human judgement. It is simply a tool that helps lawyers work more efficiently.
Has Spellbook been tested for bias?
The OpenAI and Spellbook teams are working on an ongoing basis to detect and mitigate bias in their products. While AI will never be perfect, we believe it can ultimately be trained to be less biased than the average human.
How do I get Spellbook?
Spellbook is currently in Early Access. We are gradually onboarding users that signup for our waitlist. We need to do this in order to ensure quality in our product for a broad variety of use-cases, and to ensure we have capacity to service everyone. You can sign up for the waitlist here.
Do you offer any discounts or special pricing for nonprofits?
Yes. Please reach out using the contact form above if you would like to use Spellbook to support pro-bono or nonprofit work.
How much does Spellbook cost?
We are currently exploring pricing models, but it will likely be a usage-based model, since our computing costs grow based on the amount of usage.
