WARN Page with bypass button
M
Michael Adams
Ability to create a WARN page that does not block a category/site but instead shows a warning page reminding users of some caveat before going to the requested site. For example, before going to a Generative AI site, a page appears warning the user to NOT enter any business confidential, HIPAA, or other PII into AI. There would be a button on the page to acknowledge the warning and continue to the site.
Minetta Gould
Merged in a post:
User Click-Through Override on Block Page
Ryan
I would love to see a click-through (override) option on the block page added as a configurable feature.
As an organization, our goal is not only to protect users from malicious and risky content but also to educate and empower staff to make responsible decisions.
In some scenarios we would like to warn and inform staff members rather than just blocking the website outright. As an example, for generative AI sites we could point them to our policy or guidance.
Is this a feature that could be introduced?
Minetta Gould
Merged in a post:
Feature Request: Warn-and-Continue for LLM / Generative AI Websites
C
Chris Phuah
We would like to request a new feature to support a warn-and-continue (user acknowledgement + proceed) experience for LLM / Generative AI websites, such as ChatGPT and similar tools.
The intent is to allow DNSFilter to broker access to these sites by presenting users with a custom warning or usage notice before access is granted, rather than enforcing a hard block.
In practice, when a user attempts to access an LLM site, DNSFilter would intercept the request and display a notification informing the user about acceptable use and data-handling considerations. The user would then have the option to acknowledge the notice and continue to the site. If the notice is not acknowledged, access would remain blocked.
This request follows support ticket #224564, where it was confirmed that DNSFilter currently only supports hard block pages and does not offer a warn-and-continue capability. Based on that guidance, we are submitting this as a formal feature request for future consideration by the product team.
Eric Nix
I wish there was a warn and continue for several options. (i.e., when you choose something like block new domains, block parked domains, etc., the administrator gets to choose whether to block or to warn for each category).
J
Jared Roy
I agree this would be a great feature. Shifting off of FortiNet services, I do miss the warning page. Following your example, most of our clients are allowed to use Generative AI, with some policy around how - having a warning page would be a good way to remind people that they are accessing a potentially dangerous site, and to make sure policies are followed.
To add to this, I would like to adjust the warning message per category - this way we can elaborate specifically why parked domains are warned on, or why newly observed are warned on. Some of our clients have strong-armed us into turning off blocking New domains and parked domains, despite so much AiTM coming from them.
Eric Nix
This would be a very useful page for a lot of different websites.
Jonathan Bullock
I like this idea especially for parked domains this could be nice to warn by category for why it's blocked that it's expired and at risk of being hijacked.
I foresee a few issues with this though that may need to be taken into consideration:
- For Content Delivery servers where only part of the site or modules are blocked it's not possible to display a message.
- Once a request has been answered devices would have a cached record. the initial record would need to have a very short TTL so devices would know to renew the records.
- would need to have the ability to have custom block pages per category.
- how often would you re-issue a warning? if this is for a network deployment you wouldn't know which devices were making the request unless traffic was routed through a proxy server and many sites and services have detection and block this.