Chemotaxing neutrophils and Dictyostelium amoebae produce in their plasma membranes the signaling lipid PI(3,4,5)P3 (PIP3) in gradients, which are orientated with the external chemotactic gradient and have been proposed to act as an internal compass, guiding movement of the cell. Evidence for and against this idea exists, but in all cases it depends on the use of inhibitors or gene knockouts, which may only incompletely abolish the PIP3 gradient. We have created a multiple gene-knockout strain in Dictyostelium lacking all five type-1 phosphoinositide 3-kinases encoded in the genome and the PTEN phosphatase and have thus removed all known ways for chemoattractant to produce PIP3 gradients in the plasma membrane. The resulting sextuple mutant is able to chemotax to cyclic-AMP with near wild-type efficiency and to trigger actin polymerization without apparent defect. There is, however, a consistent defect in movement speed in chemotaxis and especially in random movement. This work shows that polarization of membrane PIP3 is not necessary for accurate chemotaxis, but it can affect cell speed. A signaling pathway from receptor to cytoskeleton able to guide cells independently of polarized PIP3 and type-1 phosphoinositide 3-kinases must exist.