The point is, if investors are in at $2M, we are assuming $2M is market price. The point of market price is that includes things like "google will compete with us" or "my share will be diluted when the company raises again" as well as "the company might fail".
Now, since you personally are investing in the company (albiet with your labor rather than capital, but it's still an investment since you're paid in equity), you have to decide if it's worth investing in this particular startup. You do that by thinking about whether the startup is going to succeed or fail, since outcomes are pretty much binary.
I disagree, though, that startups have binary outcomes. In fact, one of the big problems in "investing" in startups (cash or labor) is the misalignment of incentives. It's common for the founding team to pursue outcomes that will enrich them but zero out common stock holders, and it's common for investors to push back on exits that would enrich everybody but not satisfy fund goals.
Now, since you personally are investing in the company (albiet with your labor rather than capital, but it's still an investment since you're paid in equity), you have to decide if it's worth investing in this particular startup. You do that by thinking about whether the startup is going to succeed or fail, since outcomes are pretty much binary.