“Today we sense we are close to be being able to alter human heredity,” Nobel Laureate and California Institute of Technology virologist David Baltimore said December 1 at the opening of a much-anticipated human gene editing summit taking place in Washington, D.C. this week. Gene editing, or tweaking the human genome with additions, subtractions or alterations, is becoming increasingly realistic with modern technologies. “When will we be prepared to say we are justified to use gene editing for human enhancement purposes?” he asked.

Part of the problem for researchers, doctors and ethicists alike is defining “enhancement” and deciding if it would be a move in the right direction, as the word would suggest. Is enhancement merely referring to boosting muscle tone and other desirable traits like achieving perfect pitch or does the term also encompass steps to guarantee better health by preventing disease? Some scientists disagreed over whether certain types of gene-editing would be important for helping patients, with one prominent researcher contending the technology would not often be needed, while another described dire current clinical needs for it. The international summit on gene editing, sponsored by Britain’s Royal Society, the Chinese Academy of Sciences, and the US National Academies, has grappled with such thorny questions during three days of sometimes-heated discussion on editing the human genome that comes to a close today.

Baltimore, who is chairing the summit, believes that when it comes to drawing the line for enhancement, the determining factor is whether the change would be optional versus therapeutic. “To my mind enhancement is really optional,” he says. “You are not solving a life-threatening issue.” Making a change to the PCSK9 gene, for example, would lower the risk of cardiovascular disease and for someone with high LDL—the bad kind of cholesterol – it could be the difference between life and death, he says. In that case the intervention would be therapeutic. Whether or not to engage in future “improvements” that are considered more optional, however, remains murky.

Take the DEC2 gene. Tweaking it could make a person function like the rare individuals who are born with a variant that allows them to function well with just a few hours of sleep. That trait is not necessary for most people, but it could be useful for a solider in the battlefield, for example. Ultimately, enhancements of many kinds will “definitely” happen in the future, says Fyodor Urnov of Sangamo BioSciences, a company that is working in the gene editing space. The bigger question, he says, is when it will happen.

Cheaper, and more efficient, gene editing technology that allows scientists to manipulate the human genome with greater ease and precision than ever before is forcing researchers to consider these questions quickly. Most notably, researchers are eyeing CRISPR—short for the cumbersomely-named clustered regularly interspaced short palindromic repeats. CRISPR is a powerful technology that allows editing—by way of replacing or repairing—of multiple genes at once in animal, plant and human cells. This biological dynamo could help unlock understanding of basic human biology and also help patients in need of medical care. The method has also sparked new ethical controversy. Last spring researchers in China announced they used CRISPR to alter the genomes of nonviable human embryos which could not develop into babies. They discovered the method is not yet accurate enough to be utilized in human embryos and also that it appeared to introduce unexpected mutations to other parts of the genome. Ultimately, this week’s discourse will lead to a consensus statement providing some guidance on how to approach using this and older gene editing technologies such as zinc finger nucleases and enzymes called transcription activator-like effector nucleases, or TALENs. Meanwhile, the National Academies are working on separate studies related to how to address these questions for work in other species.

At issue this week is when and how to apply gene editing for research and clinical applications in humans. Gene editing could include altering genes in one person—say to treat leukemia in one patient or make a cosmetic change—but, more controversially, it could also include making changes to the germ line that would then alter the genome for an individual’s children, grandchildren and the following generations, with potentially unknown repercussions.

Although most scientists at the meeting appear enthusiastic about conducting gene editing work to cure diseases in individual patients they remain more wary of making changes to eggs, sperm or embryos that would have lasting repercussions in future generations. The former target, say, using gene editing techniques to inactivate HIV receptors and achieve resistance of blood cells to the virus (which Sangamo BioSciences is working on in clincial trials) is different than helping parents who both carry genes for Huntington’s Disease to have a child that is free of the disease (a change to the genome that would be passed on to future generations and would likely not be very commonly needed).

For his part, Eric Lander from the Broad Institute said in a December 1 presentation at the summit that the need to employ germ line editing would remain very, very rare thanks to other already available reproductive technologies like in-vitro fertilization that could help most people.  What’s more important for avoiding genetic diseases is boosting access to gene testing so people can be more aware they are carriers for disease, he said. “What we should be thinking about is the vast majority of people with a recessive disease were not aware that they were carriers,” he said. Armed with that genetic data, people could then employ existing aides like IVF or pre-implantation genetic diagnosis to conceive healthy offspring.

Yet at the clinical level the need for such germ line editing seems more real and arguably commonplace. George Daley of Harvard Medical School said on December 1 that he and his team have seen multiple patients affected by NEMO deficiency syndrome, a disorder where an inherited faulty gene results in a weak immune system and leaves patients prone to serious infections. To care for such patients, his team has seen families that try to have a second child– sometimes nicknamed a “savior sibling”—in the hopes that the second child’s bone marrow can be employed to help the older sibling. Yet by the time families try to conceive their second child the parents are older and that likely contributes to why they have trouble getting pregnant quickly via IVF, he says. In such instances families could save time and perhaps achieve greater success if CRISPR was on the table to help ensure a single embryo with good results, he says. The question boils down to if one should consider purely the statistics or the human, says Baltimore. But either way, even if such circumstances are rare, they still exist and cannot be ignored. Perhaps the fact that CRISPR could be performed is reason enough to leave it on the table.

The need for guidance on how to attack these issues is undeniable considering the range of opinions at the conference. But guidance is not law. That may actually be a good thing.  Guidance can remain more nimble than law and it is easier to get consensus among scientists working in the area than among individuals who do not work in the field, Baltimore says. In the absence of any international body that would be an obvious fit to enforce international regulations on gene editing there are historical precedents—like stem cell research—for providing guidance and then leaving the specifics up to regional authorities. Once the consensus statement is issued, either today or in the coming weeks and months, patients, lab scientists, ethicists and medical workers will be carefully watching what comes next.