There is something here that is much more general than this particular issue: dictionary definitions of scientific concepts (and those in many other fields) are merely summaries, and far from being the last word on the matter (you can't learn quantum mechanics from a dictionary.)

It is not uncommon, in forums like HN, for there to be arguments over the 'true' definition of a word, but unless the issue being debated is lexicographical, these arguments are usually missing the point. While it is the case that the teaching of a subject often begins with some definitions, that is only the beginning, and it is usually done to shortcut the difficult process by which those definitions came to be written in the first place. On reflection, it should be clear that the definitions cannot be written (or, at least, finalized) until after we have gained an understanding of that which we are defining.

What we are observing here is the selection part of evolution. We did not observe the origination of the mutation causing the trait being acted on by selection, but it is empirically established that it is not uncommon for such mutations to occur. Nor have we seen it play out fully, where the trait, and the mutation causing it, become ubiquitous, and other alleles disappear (if the selection pressure continues and the species does not become extinct.) Here, we are observing evolution in progress, not its ultimate result.

This change alone will not create a new species, but if enough changes occur that the descendants could not, in principle, breed successfully with their ancestors, then it has. We can see a precursor to this in the mutation being lethal in males.

genetics is random. Each new generation combines genetic material from their ancestors, plus some new random mutations.

natural selection, i.e. the environment they're born in, is what picks winners and losers among those inherited characteristics and mutations.

Almost all individual mutations on a genetic level are benign. It's the combination of (selection of specific trait + compounding change) that is what we know as evolution.

In this case, the "change to the genetic code" produces elephants without tusks. That happened by change. But now, the elephants without tusks are the ones who get to live, because they're not poached. That specific mutation allows them to live/thrive. In the same way better hearing might allow a bird to evade its predators.

"no change to genetic code" ignores the fact that there are elephants wandering around who don't have (and will never have) tusks. (also: evolution is quite slow. this by any right is a huge, "forced" evolution. not in that we forced a change in the genetic code through editing it, but we are changing the course of genetics in a species through our collective behaviour).

Evolution doesn't necessarily imply "new species" (species is actually a pretty vague term) and dictionary definitions are limited.

Try this resource instead: https://evolution.berkeley.edu/evolibrary/article/evo_01

You don't have to change genes to change their presence in animals.

It's like market share. More users can simply download your software and you can gain market share without releasing an update.