Disclaimer: I know essentially nothing about US legislation, scientific ethical frameworks, etc. as I am not American. I just read the paper and have some background in genetics.
tl;dr: No, this is classic gain-of-function research as far as I can tell
From the paper, I can see two vaguely plausible arguments for why this isn't gain-of-function research:
However, neither of these arguments really holds water in my opinion. The first seems the strongest-- my main concern is that introducing a different spike protein could plausibly increase transmissibility or pathogenicity, but I don't know enough about that topic specifically to confidently evaluate that claim. If anybody does know I'd be interested to hear (for instance, do any SARS-CoV-2 variants have spike mutations?).
As to the second point; if you didn't think it was plausible that SHC014-MA15 (the chimeric virus) would be capable of infecting human cells... why did you do the test in the first place?
Some relevant things in the paper:
It's certainly "GoF-adjacent" research at least, but I don't think it's unreasonable to say that (1) it's not exactly GoF research and (2) it doesn't fall into the categories forbidden by US policy from 2014 to 2017. (The moratorium was on research "that may be reasonably anticipated to confer attributes to influenza, MERS, or SARS viruses such that the virus would have enhanced pathogenicity and/or transmissibility in mammals via the respiratory route". This research did modify a SARS virus, namely SARS-MA15, but in a way that wasn't expected to make it more harmful or more transmissible in mammals via the respiratory route; it in fact seems to have made the virus less harmful to the animals it's best suited to, namely mice, but it turned out that it did make it more able to infect humans. I guess a lot turns on what counts as "reasonably anticipated".)
[I am neither a biologist nor a lawyer.]
Do we know on what basis the NIH classified this as not GoF research? It might be worth separating out "the NIH uses unexpected/weird/just-plain-wrong criteria" from "the NIH's criteria are not being applied correctly in this case".
I don't know on what basis the NIH made that classification or when it made it. If someone knows I would be interested in reading the underlying NIH analysis.
While I could have linked to the more political claims I decided against it, to keep the politics in the question to the minimum.
From the A SARS-like cluster of circulating bat coronaviruses shows potential for human emergence by Baric, Shi et al:
To me building chimeric viruses and then infact human cells (HeLa cells are human cells) looks like dangerous gain-of-function research. Fauci seems to argue that someone the NIH is able to define this work as not being gain-of-function research. To me this redefinition seems to be the bureaucratic way they circumvent the gain-of-function moratorium. Before the moratorium was imposed Fauci was arguing against it and the moratorium wasn't imposed by anyone in the NIH or the HHS but the Office of Science and Technology Policy. To me that looks like a way to evade safety regulation by the NIH by dedefining terms because the NIH didn't like the moratorium.
This question is about more then just assigning guilt for things that happened in 2015. If we want to prevent further risk, getting the NIH to accept that growing chimeric viruses that infect human cells is what the gain-of-function regulation is supposed to prevent seems to me to be very important.
It's likely also a good case study for evading safety regulation and we should think about it in other context as well. If we end up with AI safety regulation, how do we prevent the people causing problems from just redefining the terms so that it doesn't apply to them?
If anyone has a defense of not classifying this work as gain-of-function research I'm also happy to hear that.