All the empirical evidence available says that the universe had a beginning. However, this didn’t sit well with scientists committed to a naturalistic explanation. So various attempts have been made to invent a model which avoids a beginning, doing away with any need for a creator. These attempts have so far been unsuccessful.
The BGV theorem was developed by Arvind Borde, Alan Guth and Alex Vilenkin in 2003. The conclusion of this theorem is that any mathematical model for the universe which is on average expanding, must have be time-bound in the past i.e. it must have had a beginning a finite time in the past. Alexander Vilenkin drew the following conclusion:
“It is said that an argument is what convinces reasonable men and a proof is what it takes to convince even an unreasonable man. With the proof now in place, cosmologists can no longer hide behind the possibility of a past-eternal universe. There is no escape: they have to face the problem of a cosmic beginning”
This implication has been the subject of disputes, most notably in the debate between William Lane Craig and Lawrence Krauss. Krauss claimed Craig was misinterpreting the data in favour of creatism but this was refuted by WLC in the article linked below: