This is the editorial for this year’s "Shallow Review of AI Safety". (It got long enough to stand alone.)
Epistemic status: subjective impressions plus one new graph plus 300 links.
Huge thanks to Jaeho Lee, Jaime Sevilla, and Lexin Zhou for running lots of tests pro bono and so greatly improving the main analysis.
tl;dr
* Informed people disagree about the prospects for LLM AGI – or even just what exactly was achieved this year. But they at least agree that we’re 2-20 years off (if you allow for other paradigms arising). In this piece I stick to arguments rather than reporting who thinks what.
* My view: compared to last year, AI is much more impressive but not proportionally more useful. They improved on some things they were explicitly optimised for (coding, vision, OCR, benchmarks), and did not hugely improve on everything else. Progress is thus (still!) consistent with current frontier training bringing more things in-distribution rather than generalising very far.
* Pretraining (GPT-4.5, Grok 3/4, but also the counterfactual large runs which weren’t done) disappointed people this year. It’s probably not because it wouldn’t work; it was just ~30 times more efficient to do post-training instead, on the margin. This should change, yet again, soon, if RL scales even worse.
* Edit: See this amazing comment for the hardware reasons behind this, and reasons to think that pretraining will struggle for years.
* True frontier capabilities are likely obscured by systema...
Make this part of your paper trail
Save this paper to a shelf, write a review, and keep your own notes.