Are the hyperlinks GPT-4 writes to cite a fragment of text broken atm?
I'm talking about a custom gpt, using actions.
It constructs them well when asked to simply write them down.
They are not clickable when wrote down plainly, same in markdown format.
Phone app works just fine.
#Broken hyperlinks
hardy copper
sacred helmBOT
Hi @hardy copper. I'm kapa.ai, your friendly support bot. I will try to answer your question by searching through the knowledge base. :wave:
Based on the information provided in the extracts, it appears that there have been issues with hyperlinks not being clickable in custom GPT models. This issue seems to be affecting both plain text and markdown formats.
According to the OpenAI community forum (https://community.openai.com/t/565499), users have reported that the links generated by the GPT model are not clickable. This issue seems to be reproducible across different GPT models and accounts.
However, the OpenAI help article (https://help.openai.com/en/articles/8838458-why-aren-t-my-gpt-links-clickable) provides some guidance on how to make links clickable. To make a link clickable, your custom action will need to output the full, qualified URL for the model to reference. A fully qualified URL includes all necessary components, including the scheme (http or https), domain, path, and any query parameters.
Here are the steps you can follow to ensure your links are clickable:
- Check URL format: Ensure the URL starts with "http://" or "https://".
- Validate domain and path: Confirm the domain is correctly spelled and the path accurately leads to the intended content.
- Review the custom action's URL generation: If the URL is dynamically generated, review your action to ensure it constructs the URL correctly.
If you've followed these steps and are still experiencing issues, it may be a bug that OpenAI needs to address.
Note: you can ask me a follow up question by @ mentioning me again :speech_balloon:
hardy copper
@sacred helm what is the url_safe method and why certain links are considered unsafe by the openai backend api?
sacred helmBOT
Let me check the knowledge base for you. One moment! :books:
Based on the provided extracts, OpenAI has mechanisms in place to ensure the safety of the links shared through its systems. For instance, the URL for ChatGPT shared links should always start with <https://chat.openai.com/share/> as mentioned in the OpenAI help article (https://help.openai.com/en/articles/7925741-chatgpt-shared-links-faq#h_7e06882161). Users are advised to be cautious before clicking on links and to ensure the shared link is coming from a trusted source.
In terms of the url_safe method, there is no specific information provided in the extracts. However, in general, a url_safe method or function in a programming context is typically used to ensure that a string can be safely used within a URL without causing any issues. This often involves encoding special characters and removing any potentially harmful input.
OpenAI's moderation filter (https://platform.openai.com/docs/models/moderation) is designed to detect sensitive or unsafe text coming from the API. This filter errs on the side of caution, which can result in more false positives. If you believe certain outputs have been misclassified as unsafe, you can report these using the thumbs-down icon in the Playground.
To make your application safer, you can add a moderation layer to the outputs of the Chat API by following OpenAI's moderation guide (https://platform.openai.com/docs/guides/moderation). This can help prevent content that violates OpenAI’s usage policies from being shown. OpenAI also provides a safety guide (https://platform.openai.com/docs/guides/safety-best-practices) with more information on how to build safer systems.