Overview
false does nothing and exits with status 1, which means failure. It is useful for testing how a script behaves when a step fails, or as a condition that is never true. Its partner true always succeeds.
Background
The false command exists so that a shell script can say "failure" without doing anything. Today the name covers the Bash builtin, GNU false, the false that macOS ships, and the false applet in BusyBox. They agree on exiting with a non-zero status and printing nothing.
Examples
Show that the exit status is 1:
false; echo $?
Run the || branch because the first command failed:
false || echo failed
Choose the else branch of an if:
if false; then echo a; else echo b; fi
The && branch is skipped after a failure:
false && echo skipped; echo end
Common mistakes
false ignores its arguments. Do not treat the exact number 1 as part of its contract in a portable script: test for "not zero", as if ! cmd does. In a script with set -e, a bare false stops the script, which is a handy way to see whether set -e is on. false never prints anything, so a failure is silent.
Portability
The Common examples ran in the Bash builtin, GNU false, macOS /usr/bin/false and BusyBox false with the same output. The Common examples were executed on Bash 5.2, GNU coreutils 9.11, macOS system tools, and BusyBox 1.37.0 (see the development guide for the exact check).
About BusyBox false
BusyBox bundles false as an applet, and anything beyond the common options depends on how it was built. This page's examples ran on BusyBox 1.37.0.
BusyBox notes
BusyBox false supports the options used in the Common examples. Others, and GNU long options, depend on the build.
References
Original explanations, with upstream documentation for further reading. FreeBSD references are comparison material, not a claim of Apple version equivalence.