Genetic modification of human embryos is "essential" and should be allowed so scientists can better understand basic biology, according to a new report.
The call has been made by the Hinxton Group, a worldwide network of science and ethics experts, who said gene editing of early stage embryos would be of "tremendous value" to scientific research and could have practical applications.
But the group, which met in Manchester last week, added that the technology was not yet advanced enough to be used in the reproduction process.
The report acknowledged some may find the notion of genetically modified babies "morally troubling", in particular engineered DNA which could be inherited by future generations through germline genes passed on by eggs and sperm.
Sarah Chan, a leading member and an academic at the University of Edinburgh, said preventing research in the area would be "dangerous".
She said: "Genome editing technologies hold huge potential for advancing basic research and improving human health. The prospect that genome editing may one day be used to create genetically modified humans should not in itself be cause for concern, particularly where what is at stake is curing or preventing serious disease.
"At the same time, it is clear that such applications raise more and different issues; more research is needed, together with robust public discussion, before genome editing could proceed to reproductive clinical applications.
"Restricting research because of concerns that reproductive application is premature and dangerous will ensure that it remains forever premature and dangerous, for want of better knowledge."
Earlier this year, Chinese scientists used a molecular cut-and-paste technique to edit a problem gene that causes a potentially fatal inherited blood disorder. But the move was met with calls for a worldwide ban on the creation of "designer babies".
The new report outlined possible clinical applications, including to correct mutations which cause disease or changes to prevent possible disease, but acknowledged some enhancements "may be more contentious than others".
A statement backed by the group's 22 members said: "We believe that while this technology has tremendous value to basic research and enormous potential for somatic clinical uses, it is not sufficiently developed to consider human genome editing for clinical reproductive purposes at this time.
"Given all safety, efficacy and governance needs are met, there may be morally acceptable uses of this technology in human reproduction, though further substantial discussion and debate will be required."
Senior member Robin Lovell-Badge, head of the laboratory of stem cell biology and developmental genetics at the Francis Crick Institute in London, said: "Much of our knowledge of early development comes from studies of mouse embryos, yet it is becoming clear that gene activity and even some cell types are very different in human embryos.
"Genome editing techniques could be used to ask how cell types are specified in the early embryo and the nature and importance of the genes involved. Understanding gained could lead to improvements in IVF and reduced implantation failure, using treatments that do not involve genome editing."
But the notion has met with doubt among other scientists.
Professor Emmanuelle Charpentier, one of those behind the development of the CRISPR/Cas9 DNA editing technique, told the BBC: "Personally, I don't think it is acceptable to manipulate the human germline for the purpose of changing some genetic traits that will be transmitted over generations.
"I just have a problem right now with regard to the manipulation of the human germlines."