It depends on what the tool does with the material. Writford stores client data in the UK and encrypts it in transit and at rest. AI requests are processed by Amazon Web Services and may be processed outside the UK under the UK's approved data transfer safeguards, and client files, prompts and answers are never used to train AI models. Matter access rules still govern what any individual can see.
- Is my client data used to train AI models?
- No. Client files, prompts and answers are never used as training data, and your firm's material is never pooled with another firm's. Indexing your own documents so your own fee earners get better answers is a different thing from handing them over to improve a model.
- Where is my data processed?
- Documents and matters are stored in the UK, in the AWS London region. AI requests are processed by Amazon Web Services and may be processed outside the UK under the UK's approved data transfer safeguards. Client files, prompts and answers are never used to train AI models.
- Does using AI put privilege at risk?
- Writford is built on the principle that solicitor-client privilege extends to your AI tools. Client material is processed under the data processing agreement, is not training data, and is not shared with other firms.
- Is data encrypted?
- Yes, in transit and at rest.
- Can I control who in the firm sees what?
- Yes. Four roles set access once at firm level: a fee earner sees their own matters, a supervisor and an administrator see everything, and a cashier can see every matter and edit none. Those rules apply to what the AI will surface for a given person too.
- Who are your sub-processors?
- They are listed publicly on the sub-processors page, which names every third party involved in processing and what each one does.