Recently Intel indicated their longer term goal with 1.4 nm is nominally expected circa 2029 based on current developments. 1.4 nm is definitely a lot of blue sky optimism.
Intel has been sluggish developing 10nm so moving to 7nm or 5 nm seems like a stretch. The move to 3 nm is even far remote. The move to 2 nm is even more doubtful any time soon. More than likely that once 10 nm yields improve that there will be a few years of fine tuning with 10+ nm etc.
This slide shows the roadmap for investors who want to see a path for more profits.
EUV seems to be the way for 10 nm and probably will be viable for 7 nm and even 5 nm. The synchrotron is probably going to become a common piece of equipment as this is the only source of bright light at very short wavelengths.
Intel has had yield issues with more conventional approaches while TSMC has used EUV successfully. For this reason, Intel has ordered more capital equipment for their various foundries.
Given 7 nm is slated to launch in the fourth quarter of 2021, any small delay in the next decade would let 3 nm slip into 2030 or beyond. This would more than likely be the case.