Most experts believe that P≠NP. So it is refreshing that Bollobás breaks with the conventional wisdom:
“My hunch is that P=NP, contrary to general belief.”
The quote appears in a paper included in an interesting collection on the future of graph theory, which is a proceedings from an early 90s conference. Here’s the paper and the reference: B. Bollobás, “The Future of Graph Theory”, in Quo Vadis, Graph Theory?, ed. by J. Gimbelá, J. W. Kennedy, and L. V. Quintas, North Holland, 1993, 5–11 .