Scientists can keep a liver "alive" in the laboratory for months.
But let's look at this another way.
Here's the Wiki page on life:
en.wikipedia.org
Note their criteria: homeostasis, organization, metabolism, growth, adaptation, response to stimuli, reproduction.
But note also what then amounts to an "exception" (they call it "non-cellular life"), when they talk about viruses.
So then, how would we classify a prion, which they don't talk about? A prion is a misfolded protein, which when it interacts with other like proteins, causes them to misfold too.
en.wikipedia.org
It satisfies all the criteria of life, it reproduces, it's organized, it has a rudimentary form of homeostasis and response to stimuli insofar as its folding protects itself from intracellular degradation, and whenever it contacts a like protein it "transmits" something (even if it's only information).
The only difference between a protein and a rock is that carbon atoms allow a symmetry that supports complexity. We call it "organic" chemistry. But there are certainly inorganic preparations that support complexity - for example, ferromagnetism, which when it interacts with external magnetism can solve very complex computational problems. (See 'Ising model'). The information in a partitioned spin glass can reproduce itself across partitions - it responds to stimuli, it's organized, and it's homeostatic because it obeys a Hamiltonian.
There is no clearly defined line of demarcation.
How about a robot made of metal and chips, that can build another robot just like itself? Is it alive?
How about an AI without a body, "pure consciousness" - is it alive? I think therefore I am?
All "life" really is, is a complex state of matter. It's the complexity that distinguishes what we ordinarily refer to as "life". In defiance of entropy, life packs a whole lot of information into a very tiny space. Chemically, and in ferromagnetism and such, the complexity is made stable by "relationships at a distance", what are euphemistically called "coupling constants" (as in the Belousov-Zhabotinsky reaction in simple form, and a ferromagnetic spin glass in more complex form). In the latter case, the arrangement of coupling constants closely resembles the synaptic weights in a neural network.
"Relationships at a distance" is exactly what entanglement is.