It’s funny, because while technically correct, you may end up fundamentally wrong that there was a limit to the value we might find in parallel compute. After all, we may have lie of sight to AGI through parallel scale alone.
Of course there may be some practical limit to Gustafson's Law, but I don't think we've found it yet, at least in many scientific domains.