Licenses get to set terms of redistribution. But training of the ML model -- the thing described by your #3 -- is not redistribution (imho). So maybe it's as unenforceable as saying left-handed people can't read your code.

The redistribution happens later, either when copilot blurps out some of your code, or when the copilot user then distributes something using that code (I'm curious which). At that point, whether some use of your code is infringing your license doesn't depend on the path the code took, does it? (in which case #3 is moot)

The BSD license also controls "use", not just "redistribution":

  Redistribution and use in source and binary forms, with or without
  modification, are permitted provided that the following conditions
  are met:

That's word-for-word BSD license.

The only change I made is adding clause 3:

  3. Use in source or binary forms for the construction or operation
     of predictive software generation systems is prohibited.

Okay; thanks for clarifying. I actually hadn't noticed that use of "use" in the BSD license before. I think I need an IP lawyer explain what that "use" means.

Many licenses have constraints, whether this wording is the best way to do it is up for discussion, but it's certainly possible to do it.