All of us are like this. There are deep reading technics that help slightly but unless you use said method in your day-to-day, you are going to start forgetting things eventually.

In fairness, offers and compensation packages have also scaled massively, especially for those research-heavy roles you mentioned so the publication requirements are not too unexpected.

They should but many don’t because often their results are not statistically significant or they have to spend a ton of compute to only show very small statistically significant improvements. So, they’ll just put 5 run averages (sometimes even less) and hope for the best. I have been a reviewer on most of the top ML conferences and I’m usually the only reviewer holding people accountable on statistical significance of results when confidence intervals are missing.

I mean it’s all relative, and in many cases where ML-based systems are saving use of the alternative solutions that cost a lot more energy, no ML is helping the environment. As it stands though, yes, many of the areas are incredibly energy-consuming.

I mean there are plenty of major areas in ML that LLMs cannot even begin to address (e.g. processing time-series data - XGBoost still reigns supreme, edge ML, etc.). Also, keep in mind that most of the people at major LLM groups are PhD so chances are if you wanna work even on LLMs, having a PhD will help. Afterall, scaling is good but if your research shows more efficient training pathways, the difference can be 9-figure sums for these companies.

There is a fair chance that a lot of LLMs already do this, just not only the test set but also other data.

My father worked all through his PhD, and by the time he finished in his early 30s, he was at the director-level in the company. However, that seems to be increasingly a rare occurrence as when I tried to do it during my master’s, my supervisor nearly dropped me and forced me into basically just working at his lab. So keep this in mind if you wanna work during your graduate degree (and find the right advisor/group in that regard).

I also think a large point of value in doing a PhD is publishing a lot of cutting-edge research. To me, a master’s with many papers probably positions you better for the jobs you noted than a PhD without any. Lastly, if you are talking about a PhD in ML, admission to top-tier program is ridiculously competitive; so you may wanna factor that in too.