IP Author is a brand of Dolcera, a leading IP analytics company.
IP Author is a company that leverages AI technology to assist with drafting patents. One of the most reliable and efficient large language models (LLM) suitable for this purpose is ChatGPT (GPT-4). Therefore, IP Author employs ChatGPT to aid in the process of drafting patent applications. As new language models and techniques are developed, IP Author may incorporate them, provided they ensure the safety and confidentiality of customer data.
When using ChatGPT without proper instructions or data, it can do an acceptable job of drafting patent applications. However, it may also generate false references to prior art if prompted. On the other hand, IP Author overcomes all the limitations of ChatGPT by providing the following features:
- Integration with the Dolcera PCS patent search engine to obtain accurate and up-to-date patent information from around the world for prior art searching.
- Integration with various data sets, including patent data from Dolcera PCS, technology standards data, scientific literature, product information, and more.
- Use of templates approved by the user to create high-quality patent drafts for different regions.
- Creation of diagrams such as flowcharts and block diagrams using a range of tools.
Our platform IP Author uses ChatGPT via its Application Programming Interface (API). It is important to note that this service is not offered for free by OpenAI, the company that developed ChatGPT. To use OpenAI's API, we must adhere to their terms and conditions, which can be found at https://openai.com/policies/api-data-usage-policies.
- OpenAI guarantees that any data submitted by clients through their API will not be utilized to train or enhance their models unless explicitly permitted by the client.
- Additionally, OpenAI will retain all data sent through the API for a maximum of 30 days to monitor for any misuse or abuse, after which it will be deleted (unless required by law).
- Clients own the input and output of the models, as stated in the API agreement which is available at https://openai.com/policies/terms-of-use.
IP Author has extensive security controls in place to protect client confidential information and to share the results of IP Author's output securely with our users.
IP Author has extensive security controls in place to protect client confidential information and to share the results of IP Author's output securely with our users.
Authentication and Authorization
- A limited number of authorized Dolcera employees, who are subject to confidentiality and security obligations, can access this data to run analytics (e.g. the use of certain product features). Dolcera regularly reviews and updates the authorizations of team members as appropriate, based on their work assignments.
Infrastructure logs and audit trails contain information about security-related events including logins, IP address, date, and time of access.
Physical Security
- US data center facilities are protected by the highest level of physical and biometric access controls.
Redundancy
- Dolcera systems have several levels of redundancy, including multiple servers, multiple storage and backup solutions, multiple network connections, and multiple levels of physical and data security.
Data Backups
- Data is backed up on a nightly basis or in real-time as appropriate and is securely synchronized to the Dolcera servers located in the US data center.
Intrusion Detection
- Intrusion detection systems have been installed on Dolcera servers and are monitored by the Dolcera team.
Disaster Recovery
- Dolcera has a disaster recovery plan and the necessary technology and systems (including data backups and alternative designated work sites) to implement the disaster recovery procedures in case of need.
See above
The API Agreement for OpenAI is available here: https://openai.com/policies/terms-of-use
- OpenAI will only use the user data to answer user requests, and (if necessary) for abuse prevention
- IP Author will only use the user data to provide the IP Author service, and for high-level analytics (e.g. to measure the frequency of use of certain IP Author features)