The report has three main recommendations: all areas of research must be given the opportunity to thrive; technology-based industries should be given the chance to grow; and education must be strengthened at all levels. To encourage research, the report states that the government must "make federal research funding stable and substantial, maintain diversity in the federal research portfolio, and promote creative groundbreaking research."

In writing the report the committee was guided by the vision that "the USA must maintain and improve its pre-eminent position in science and technology in order to advance human understanding of the universe and all it contains, and to improve the lives, health, and freedom of all peoples."

The report also contains some of the evidence presented by scientists to the committee. James Langer, a physicist from the University of California at Santa Barbara, explained how difficult it was to know which research projects were going to be successful or have the greatest impact. "Newton spent a large part of his career studying alchemy, " Langer told the committee. "Einstein devoted the second half of his life to problems that we now know could not be solved without modern discoveries in elementary particle physics. Bardeen grossly underestimated the importance of his invention of the transistor, as did major US corporations at the time. I am equally certain that we cannot trust scientists, engineers, or public policy experts to predict where those advances will occur or in what ways they will have there greatest impacts."

The report also considers how peer review can stifle creativity and risk-taking in research. "There are no rewards for risky science, " Suzanne Rutherford of the University of Chicago told the committee. "It is too important to publish." The report recommends that a fraction of government research grants should be reserved for speculative research, although the money should still be allocated through peer review.

And in an effort to make the outcome of government-funded research more understandable, the report recommends that plain English summaries of all results, including their implications, should be publicly available on the Internet.