I' ve been in that situation dozen's of time and it is usually a case where it is almost impossible to make an estimate because the stakeholders have not articulated what they want sufficiently. The worst managers will say "go back and think about it - sharpen your pencil". The best will respond, "I understand you can't give you an estimate until you understand more about the project. Let me see what I can do to help you."

An even more precarious situation is deadline driven projects where the work is not yet identified. In this case the boss will ask "will we meet the deadline?" Don't answer that question - estimating unknown work is a trap and estimates are useless until you start the discovery process and assign work - if you have a hard and fast deadline there is only one approach possible - triangulation - constantly measure your progress and determine if the deadline is achievable - use a dashboard that is easy for management to understand that shows the current state of the project green, yellow, red.

Be realistic, not optimistic. While I generally believe it is "always too early to panic", deadline driven projects are an exception - early communication of risks allows everyone involved to help mitigate those risks. More projects fail because of risks that were not identified or communicated early enough than one can count.