Some gamma-ray burst (GRB) spectra exhibit high-energy tails with the
highest photon energy detected at 18 GeV. The spectral slope of the hi
gh-energy tails is sufficiently flat in vF(v) to consider the possibil
ity of their detection at still higher energies. We calculate how many
bursts can reasonably be expected above a given energy threshold for
a cosmological distribution of bursts satisfying the observed apparent
brightness distribution. The crucial point is that the gamma-ray abso
rption by pair production in the intergalactic diffuse radiation field
eliminates bursts from beyond the gamma-ray horizon tau(gamma gamma)
similar to 1, thus drastically reducing the number of bursts at high e
nergies. Our results are consistent with the nondetection of bursts by
current experiments in the 100 GeV to 100 TeV energy range, For the e
arthbound detector array MILAGRO, we predict a maximal GRB rate of sim
ilar to 10 events per year. The Whipple Observatory can detect, under
favorable conditions, similar to 1 event per year. The event rate for
the HEGRA array is similar to 0.01 per year. Detection of significantl
y higher rates of bursts would severely challenge cosmological burst s
cenarios.