There are chains that currently use a slot time of 12s and 6s (async backing). It is important to find out what would break and requires any storage migration if we would increase the slot to 24s here:

More context:

• Assuming reference hardware and full usage of cores, for every block author change, a 3 core elastic scaling chain (using AURA) will loose 2s of execution at the beginning of its slot. This is because the next author has to import the best block built by previous collator before it's slot starts.
• If we increase the slot to 24s this will improve throughput while also preserving censorship resistance.