The bun acquisition is driven by current AI capabilities.
This argument requires us to believe that AI will just asymptote and not get materially better.
Five years from now, I don't think anyone will make these kinds of acquisitions anymore.
> This argument requires us to believe that AI will just asymptote and not get materially better.
That's not what asymptote means. Presumably what you mean is the curve levelling off, which it already is.
> This argument requires us to believe that AI will just asymptote and not get materially better.
It hasn't gotten materially better in the last three years. Why would it do so in the next three or five years?
An Anthropic engineer was getting some attention for saying six months: https://www.reddit.com/r/ClaudeAI/comments/1p771rb/anthropic...
I assume this is at least partially a response to that. They wouldn't buy a company now if it would actually happen that fast.