Seems like it was just getting distracted? Happens to me all the time, I stare into space etc. not sure what your point is. No one is claiming GPT-5 is AGI, and also, these sorts of behaviors are compatible with AGI anyway, so they don’t really provide much evidence one way or another.
I would’ve mapped “distracted” to “starts to think about something else”, not “thinks about nothing”. If mapping to human behavior, this is IMO more like “mindlessly staring at a wall under mental fatigue” or something.
More seriously, what this actually immediately reminded me of is this paper:
Let’s Think Dot by Dot: Hidden Computation in Transformer Language Models
Chain-of-thought responses from language models improve performance across most benchmarks. However, it remains unclear to what extent these performance gains can be attributed to human-like task decomposition or simply the greater computation that additional tokens allow. We show that transformers can use meaningless filler tokens (e.g., ‘......’) in place of a chain of thought to solve two hard algorithmic tasks they could not solve when responding without intermediate tokens. However, we find empirically that learning to use filler tokens is difficult and requires specific, dense supervision to converge. We also provide a theoretical characterization of the class of problems where filler tokens are useful in terms of the quantifier depth of a first-order formula. For problems satisfying this characterization, chain-of-thought tokens need not provide information about the intermediate computational steps involved in multi-token computations. In summary, our results show that additional tokens can provide computational benefits independent of token choice. The fact that intermediate tokens can act as filler tokens raises concerns about large language models engaging in unauditable, hidden computations that are increasingly detached from the observed chain-of-thought tokens.
I expect it’s not one of the cases where the filler tokens can encode useful computations, but there may actually be some semantic meaning there.
No one is claiming GPT-5 is AGI, and also, these sorts of behaviors are compatible with AGI anyway, so they don’t really provide much evidence one way or another.
(That was intended as a tongue-in-check joke implying that executive-function problems are what qualifies a system as AGI.)
Seems like it was just getting distracted? Happens to me all the time, I stare into space etc. not sure what your point is. No one is claiming GPT-5 is AGI, and also, these sorts of behaviors are compatible with AGI anyway, so they don’t really provide much evidence one way or another.
I would’ve mapped “distracted” to “starts to think about something else”, not “thinks about nothing”. If mapping to human behavior, this is IMO more like “mindlessly staring at a wall under mental fatigue” or something.
More seriously, what this actually immediately reminded me of is this paper:
I expect it’s not one of the cases where the filler tokens can encode useful computations, but there may actually be some semantic meaning there.
(That was intended as a tongue-in-check joke implying that executive-function problems are what qualifies a system as AGI.)