FYI -  https://github.com/WICG/autofill-event/pull/30

Any other feedback would be appreciated! :)

On Tue, Jan 27, 2026 at 2:06 PM Yoav Weiss (@Shopify) <
[email protected]> wrote:

> Hey Chris!
>
> On Mon, Jan 26, 2026 at 9:10 PM Chris Thompson <[email protected]> wrote:
>
>> Hi Yoav --
>>
>> Chrome web platform security folks took a quick look at this -- we think
>> there aren't any concerns but I think the spec could be clarified a bit to
>> make it more obviously the case :-)
>>
>> Section 6 flags that UAs SHOULD only expose the data after explicit user
>> confirmation. We think this is also important irrespective of timing
>> attacks.
>>
>
> Yeah, the "timing attack" title is not appropriate. I'll revamp.
>
>
>> It also wasn't immediately clear to us that the data being passed is
>> limited to the specific autofill fields/type
>> <https://wicg.github.io/autofill-event/#dom-autofillevent-autofillevent-type-eventinitdict-type>
>> initially requested (and thus confirmed by the user) and this *cannot* be
>> changed by the page handling the autofill event -- i.e., that all that can
>> happen is a refill for the same data the user already consented to share.
>> Maybe this is just under-specified and UA-dependent?
>>
>
> It's indeed not specified (as it's not specified today), but at the same
> time, the spec does define the "full-address" autocomplete value
> <https://wicg.github.io/autofill-event/#full-address> that enables the UA
> to know ahead of time that it needs to get user consent for the entire
> address information, not just the fields currently present in the form.
>
> We heard feedback from some UAs that this would be required for them,
> while others felt that getting user consent of the specific parts of the
> address implies consent for the more generic parts.
>
> The attribute will enable each UA to apply the policy it deems fit.
>
> I think it would still be good to at least discuss in the Security/Privacy
>> considerations section, even if it's non-normative.
>>
>
> Sure!
>
>
>> Additionally, given that requirement, the new "full-address" field name
>> may have UX challenges to ensure that the user understands the full scope
>> of what is being shared.
>>
>
> Potentially, but I imagined it being equivalent to pretending that all the
> fields that are relevant to the user's address are already present in the
> form.
>
>
>>
>> Cheers,
>> Chris
>>
>> On Wednesday, January 14, 2026 at 6:22:57 AM UTC-8 Yoav Weiss (@Shopify)
>> wrote:
>>
>>> *Contact emails*
>>> [email protected], [email protected]
>>>
>>> *Explainer*
>>> https://github.com/WICG/autofill-event?tab=readme-ov-file#autofill-event
>>>
>>> *Specification*
>>> https://wicg.github.io/autofill-event
>>>
>>> *Summary*
>>> Autofill is a key feature of the web that reduces friction for millions
>>> of users everyday. But getting autofill to work reliably with dynamic forms
>>> across multiple implementations requires jumping through many hoops. This
>>> feature adds an "autofill" event that would allow developers to modify
>>> their forms to fit the autofilled data and let the browser know when they
>>> have done so.
>>>
>>> *Blink component*
>>> Blink
>>> <https://issues.chromium.org/issues?q=customfield1222907:%22Blink%22>
>>>
>>> *Web Feature ID*
>>> Missing feature
>>>
>>> *Motivation*
>>> Autofill today relies on static forms, but address forms are dynamic,
>>> almost by definition. Addresses around the world have different fields and
>>> requirements, and developers have to reconcile that dynamic nature with
>>> autofill's static nature. Some browser heuristics are simplifying that task
>>> in some implementations, but dealing with that implementation-specific
>>> complexity requires jumping through a bunch of hoops. This feature will
>>> enable developers to be notified of autofill-filled values, modify the form
>>> to match the filled data and let the browser know when that modification
>>> happened, enabling the browser to refill the data in the adapted form.
>>>
>>> *Initial public proposal*
>>> https://github.com/WICG/proposals/issues/162
>>>
>>> *Requires code in //chrome?*
>>> True
>>>
>>> *Tracking bug*
>>> https://issues.chromium.org/issues/466333215
>>>
>>> *Estimated milestones*
>>>
>>> No milestones specified
>>>
>>>
>>> *Link to entry on the Chrome Platform Status*
>>> https://chromestatus.com/feature/5137581018841088?gate=5148847959572480
>>>
>>> This intent message was generated by Chrome Platform Status
>>> <https://chromestatus.com/>.
>>>
>>

-- 
You received this message because you are subscribed to the Google Groups 
"blink-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion visit 
https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAOmohSKqgEjSSSHu2Wcbhvx8-8SwzFRui2ztMF8CJ%3DvC%2BfLcQw%40mail.gmail.com.

Reply via email to