651–660 of 1001 posts
Not so sure about the behind the scenes "automatic router". What's to stop OpenAI from slowing gimping GPT-5 over time or during times of high demand? It seems ripe for delivering inconsistent results while not changing the price.
It is frequently suggested that once one of the AI companies reaches an AGI threshold, they will take off ahead of the rest. It's interesting to note that at least so far, the trend has been the opposite: as time goes on and the models get better, the performance of the different company's gets clustered closer together. Right now GPT-5, Claude Opus, Grok 4, Gemini 2.5 Pro all seem quite good across the board (ie the…
SGI would be self-improving to some function with a shape close to linear based on the amount of time & resources. That's almost exclusively dependent on the software design, as currently transformers have shown to hit a wall at logarithmic progression x resources.
In other words, no, it has little to do with the commercial race.
It is frequently suggested that once one of the AI companies reaches an AGI threshold, they will take off ahead of the rest. It's interesting to note that at least so far, the trend has been the opposite: as time goes on and the models get better, the performance of the different company's gets clustered closer together. Right now GPT-5, Claude Opus, Grok 4, Gemini 2.5 Pro all seem quite good across the board (ie the…
I have a had a bunch of positive experiences as well, but when it goes bad, it goes so horribly bad and off the rails.
It is frequently suggested that once one of the AI companies reaches an AGI threshold, they will take off ahead of the rest. It's interesting to note that at least so far, the trend has been the opposite: as time goes on and the models get better, the performance of the different company's gets clustered closer together. Right now GPT-5, Claude Opus, Grok 4, Gemini 2.5 Pro all seem quite good across the board (ie the…
It's also worth considering that past some threshold, it may be very difficult for us as users to discern which model is better. I don't think thats what's going on here, but we should be ready for it. For example, if you are an ELO 1000 chess player would you yourself be able to tell if Magnus Carlson or another grandmaster were better by playing them individually? To the extent that our AGI/SI metrics are based on…
An average driver evaluating both would have a very hard time finding the f1s superior utility
It is frequently suggested that once one of the AI companies reaches an AGI threshold, they will take off ahead of the rest. It's interesting to note that at least so far, the trend has been the opposite: as time goes on and the models get better, the performance of the different company's gets clustered closer together. Right now GPT-5, Claude Opus, Grok 4, Gemini 2.5 Pro all seem quite good across the board (ie the…
It's also worth considering that past some threshold, it may be very difficult for us as users to discern which model is better. I don't think thats what's going on here, but we should be ready for it. For example, if you are an ELO 1000 chess player would you yourself be able to tell if Magnus Carlson or another grandmaster were better by playing them individually? To the extent that our AGI/SI metrics are based on…
one can intentionally use a recent and a much older model to figure out if the tests are reliable, and in which domains it is reliable.
one can compute a models joint probability for a sequence and compare how likely each model finds the same sequence.
we could ask both to start talking about a subject, but alternatingly each can emit a token. look again at how the dumber and smarter models judge the resulting sentence does the smart one tend to pull up the quality of the resulting text, or does it tend to get dragged down more towards the dumber participant?
given enough such tests to "identify the dummy vs smart one" and verifying them on common agreement (as an extreme word2vec vs transformer) to assess the quality of the test, regardless of domain.
on the assumption that such or similar tests allow us to indicate the smarter one, i.e. assuming we find plenty such tests, we can demand model makers publish open weights so that we can publically verify performance agreements.
Another idea is self-consistency tests: a single forward inference of context size say 2048 tokens (just an example) is effectively predicting the conditional 2-gram, 3-gram, 4-gram probabilities on the input tokens. so each output token distribution is predicted on the preceding inputs, so there are 2048 input tokens and 2048 output tokens, the position 1 output token is the predicted token vector (logit vector really) that is estimated to follow the given position 1 input vector, and the position 2 output vector is the prediction following the first 2 input vectors etc. and the last vector is the predicted next token following all the 2048 input tokens. p(t_(i+1) | t_1 =a, t_2=b, ..., t_i=z).
But that is just one way the next token can be predicted using the network: another approach would be to use RMAD gradient descent, but keeping model weights fixed, and only considering the last say 512 input vectors as variable, how well did the last 512 predicted forward prediction output vectors match the gradient descent best joint probability output vectors?
This could be added as a loss term during training as well, as a form of regularization, which turns it into a kind of Energy Based Model roughly.
Going by the system card at: https://openai.com/index/gpt-5-system-card/ > GPT‑5 is a unified system . . . OK > . . . with a smart and fast model that answers most questions, a deeper reasoning model for harder problems, and a real-time router that quickly decides which model to use based on conversation type, complexity, tool needs, and explicit intent (for example, if you say “think hard about this” in the prompt).…
I know this is just arguing semantics, but wouldn't you call it a unified system since it has a single interface that automatically interacts with different components? It's not a unified model , but it seems correct to call it a unified system .
It is frequently suggested that once one of the AI companies reaches an AGI threshold, they will take off ahead of the rest. It's interesting to note that at least so far, the trend has been the opposite: as time goes on and the models get better, the performance of the different company's gets clustered closer together. Right now GPT-5, Claude Opus, Grok 4, Gemini 2.5 Pro all seem quite good across the board (ie the…
The reason AGI would create a singularity is because of its ability to self learn. Presently we are still a long way from that. In my opinion we at least are as far away from AGI as 1970s mainframes were from LLMs. I really don’t expect to see AGI in my lifetime.
I'm not really convinced, the benchmark blunder was really strange but the demos were quite underwhelming, and it appears this was reflected by a huge market correction in the betting markets as to who will have the best AI by end of the year. What excites me now is that Gemini 3.0 or some answer from Google is coming soon and that will be the one I will actually end up using. It seems like the last mover in the LLM…
I really don't want the already trillion dollar mega monopoly to own the world.
It's a really good model from my testing so far. You can see the difference in how it tries to use tools to the greatest extent when answering a question, especially compared to 4.1 and o3. In this example it used 6! tool calls in the first response to try and collect as much info as possible. https://promptslice.com/share/b-2ap_rfjeJgIQsG
It is frequently suggested that once one of the AI companies reaches an AGI threshold, they will take off ahead of the rest. It's interesting to note that at least so far, the trend has been the opposite: as time goes on and the models get better, the performance of the different company's gets clustered closer together. Right now GPT-5, Claude Opus, Grok 4, Gemini 2.5 Pro all seem quite good across the board (ie the…