> we don't have a definition nor a test for consciousness
Isn't this a bit like saying we don't have a definition for "cloud" due to not having defined exactly at which density, size, etc water vapor qualifies as a cloud? There's certainly a gradient of chemical responses to outside stimulus within an organism, ranging from reflex to subconscious responses (e.g. heartbeat speed) to awareness to consciousness. Defining consciousness within that gradient would then simply consist of specifying exactly which subset of responses qualify as falling under the umbrella definition for consciousness (e.g. certain patterns of chemical response in specific parts of a brain).
There's then the argument that it's effectively impossible for two things to be exactly the same physically, so how can we know that "my" experience of consciousness is not some unique snowflake in some universe simulation. An interesting psychological tidbit that seems relevant is the observation that every salmon has the same number of bones. What's interesting about it is that many people have the incorrect intuition that skeletons within a fish species have a much higher variance than they actually do. Similarly, there's this intuition that just because neurons connect in seemingly random ways, that a more complex pattern built on top of of said neurons might then have higher variance than genetics would suggest.
The problem with a lot of philosophical arguments is that they jump from "consciousness is vaguely defined" to "therefore it's technically possible for some highly unlikely setup to be true". That's not how we do science elsewhere (i.e. we typically look at what's statistically more likely via p values, phenotypes, etc).
Another point of contention is the somewhat tautological nature of defining consciousness. It basically boils down to "it is what it is and it isn't what it isn't". But the thing is: many things are defined in tautological ways (e.g. the cloud above, or what is the first "chicken" in the "chicken and egg" problem). You could easily get out of the problem by simply specifying an arbitrary definition with arbitrary breadth, since it is already inherently vague and arbitrary to begin with. This is more or less the same process by which we came up with the definition of even hard science things like the meter.
Regardless, even without an exact definition, it's possible to engage in the discussion of ethics of lab grown consciousness, specifically how it applies to real life. One can argue that lab grown consciousness is not necessarily bad in and of itself (similar to the fact that animal slaughter exists for the purpose of feeding people). I think people objections to it stem from ascribing human rights (e.g. not being subject to cruelty) to things that are inherently similar to humans. But in that sense, it doesn't even need to be remotely close to being human-like, as we already debate about the ethics of animal torture and hypothetical scenarios that never actually play out in real life w/ real functioning people.
At the end of the day, ethics are just a framework to formulate things like rule of law, and ultimately are about a balance between individual freedom vs collective interest. We know from politics that there are virtually endless disagreements about where said balance lies. One could then argue that it's phenotypical that individuals may have disagreeing opinions about various subjects and that there's no point in pursuing "absolute righteousness" when said endeavor boils down to equally pleasing all members of a phenotypically heterogeneous population.
From there, there are two possible logical conclusions: a) that "bad" things must necessarily exist because badness is not a consensus (and contradictory depending on who you ask) and that existence is an eternal struggle or b) that the pursuit of absolute righteousness involves taking the zen approach of letting go and letting things be as they are.