UL 217 and 268 standards. They have gotten better but there are certain things where the balance of safety, cost and convenience heavily favors the first followed by the second and then the third.

Interestingly (for a certain narrow definition of "interesting"), the standards have been updated in recent years to reflect the impact of convenience on safety. Cooking can produce smoke that reaches the threshold needed to trigger the alarm without being an indication that there's an imminent danger to the structure and residents. When cooking sets off alarms, people may be tempted to disable those alarms while cooking. Some get re-connected afterward, some don't. The result is that because the alarm is effective (annoying), a significant number of homes have fewer active warning devices than intended.

Some nerds conned a lab into letting them light stuff on fire for money without all the legal trouble a casual arsonist might have to worry about. A side benefit of this arrangement is they've collected a ton of data on smoke and fire development with a wide variety of fuels over time. In order to cut down on the likelihood of those annoyance disconnections, devices built to the newest standard should be less sensitive to the type of smoke that results from normal cooking.

Doesn't fix the midnight chirp you're talking about but the people writing the requirements have noticed that human nature is still a factor.