As president, Trump will shape the future of science. And scientists are worried.

Last October, on a conservative talk radio show, Donald Trump said this: “I hear so much about the [National Institutes of Health], and it's terrible.”

Think about that for a second. The National Institutes of Health spends $32.3 billion a year on medical research. It saves lives. Its researchers have vastly advanced our understanding of diseases like cancer and diabetes.

Though there is much that is uncertain about how Trump, president-elect of the United States, will govern and what it will mean for the future of our most critical science issues, he has a long history of casting doubt and ginning up conspiracy theories about scientific facts and institutions. He’s called climate change a hoax perpetrated by the Chinese. He’s suggested he believes the thoroughly debunked link between vaccines and autism. He’s selected a noted climate change denier to lead the transition of the Environmental Protection Agency (which he has said he might abolish). He wants the US to withdraw from the Paris climate deal, which he can effectively do.

“He seems like he’ll be a pretty clearly anti-science president,” says Fred Guterl, executive editor of Scientific American, which ran a scathing editorial against Trump. “Perhaps our first.” As the prestigious scientific journal Nature put it, “Trump is a demagogue not fit for high office, or for the responsibilities that come with it.”

That’s gravely concerning, considering how much power the president has to direct health and science policy in America.

He’ll get to choose a secretary of health and human services and a director of the Centers for Disease Control and Prevention. And with his party controlling both houses of Congress, he could slash federal funding for the National Institutes of Health, the National Science Foundation, and other major government science institutions if he wants to. His disdain for science has also extended to NASA, which he called a “logistics agency for low-Earth orbit activity.”

President Obama has made significant investments (despite stagnant funding overall) in biomedical health science, like in his “moonshot” initiatives in cancer research and mapping the human brain. Trump could dismantle those.

Trump’s vice president, Mike Pence, has a grievous track record on science as well. Pence doubted that smoking causes cancer as late as 2000, is skeptical of climate change, and has said schools should teach the “controversy” around evolution.

And as Nature points out, Trump’s hard-line immigration policies “could dissuade talented foreign scientists from working or studying at US institutions.”

But it’s not the policy shifts that most worry some scientists — it’s Trump’s generally unscientific pattern of thinking.

“This is a man that has shown little interest in trying to understand science,” Paul Offit, a vaccine researcher at Children’s Hospital of Philadelphia, says. “He appears to be a science denialist.”

It should worry scientists, Offit says, that Trump puts more weight into an anecdote about a child developing autism than the overwhelming scientific consensus that vaccines do not cause autism.

Or how Trump didn’t want US Ebola victims transported to the United States for treatment, even though the victims would have died and the risk for Ebola spread in the US was minimal. Or he called for a Syrian refugee ban to the United States after the terrorist attacks in Paris last November, despite any evidence that the Paris attackers had come from Syria. Or how he resisted updating his doubts that President Obama was born outside of the United States, despite the incontrovertible evidence (Obama released his birth certificate). Trump often thinks with mental shortcuts, not data.

That mode of thinking is “worrisome for scientists who live in the world of reason and facts,” Offit says.

For scientists, the ascension of Trump raises questions about how he will compare with George W. Bush, who got generally low marks on topics like global warming and reproductive health.

Still, any major political transition is liable to bring about significant change in public science policy. As Kevin B. Marvel, the executive officer of the American Astronomical Society, points out, science research is pretty much always in peril since public science research budgets are subject to the whims of Congress. “No portion of the federal budget for discretionary expenditure is immune from decrease, increase, or even elimination, no matter what administration is in power, and scientific research … holds no special position,” he says.

And Trump could support some pro-science policies. He’s said he’d like NASA to have a renewed mission for space exploration. (Though Robert Walker, the Trump campaign’s space policy adviser, has said a Trump administration would shift NASA funds away from Earth and atmospheric science.)

And it’s possible that scientific funding, overall, will not suffer too much. In September, Trump told sciencedebate.org that “scientific advances do require long-term investment.” Trump also takes council from Newt Gingrich, who has advocated for expanding the budget of the NIH.

In the meantime, Trump will begin picking his Cabinet. The American Association for the Advancement of Science CEO Rush Holt said in a statement that he hoped Trump would appoint a “respected scientists or engineer” to serve as his science adviser.

But when it comes to selecting of advisers, well, he tends to choose people who are loyal over people who are the most qualified.