Using AI safely at work
What teams handling confidential information should know
Using generative AI at work brings pitfalls that performance and features alone cannot prevent. This series explains, without jargon, how to handle documents, decide which information AI may receive, and provide evidence for security in workplaces dealing with confidential information.
Articles in this series
Is that blacked-out text really safe?
A black rectangle in a PDF may not remove the text beneath it. Learn how to check and protect documents before publication.
Is it safe to give AI your designs and customer lists?
Explore inputs, retrieval and training, and how to handle trade secrets and personal information. Review your practices with eight questions.
Why can on-premises AI be safer?
Learn how to control the routes by which data can leave, and which security responsibilities remain in an on-premises deployment.
Related series
What is Sovereign GaiXer?
What is Sovereign GaiXer? A complete product guide using a smartphone analogy
A plain-language product guide answering “What can it actually do?” Read it to understand the system behind secure AI use.
AI and Money
AI and Money: a complete introduction to AI pricing and costs (23 parts)
A plain-language introduction answering “How much does AI actually cost?” Learn the pricing principles and decision criteria before deployment.
Contact us to discuss how these principles apply to your organization.
This series explains general principles and is not legal advice. Follow your organization’s policies and obtain professional review for individual decisions. Legal examples concern Japan unless stated otherwise.