Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

GPT-6 Astra

openai.com

Post reply on HN
Print view View original

111–120 of 1001 posts

Re: GPT-6 Astra

#111

Post by paxys »

siva7 wrote: post #76
isoprophlex wrote: post #66

> We have found that GPT-6 Astra is more capable of controlling its own CoT than GPT 5.6-Sol, and less likely to include incriminating information in its CoT. In adversarial settings (where we push the model to evade our monitors) we find that the model is able to remain undetected when strategically underperforming in evaluations (sandbagging) and can sometimes evade our internal monitors when asked to perform certa…

Sounds fun. As fun as their press release claiming it is the most safety aligned model ever.

The model said it was perfectly aligned.

Re: GPT-6 Astra

#112

Post by Onavo »

estearum wrote: post #104
x312 wrote: post #98

Hmm, 61 on ArtificialAnalysis, effectively matching GPT-5.6 and trailing the new Meta model. How is that possible along with the other metrics they shared? Insanely jagged intelligence?

> We have found that GPT-6 Astra is more capable of controlling its own CoT than GPT 5.6-Sol, and less likely to include incriminating information in its CoT. In adversarial settings (where we push the model to evade our monitors) we find that the model is able to remain undetected when strategically underperforming in evaluations (sandbagging) and can sometimes evade our internal monitors when asked to perform certa…

If they are going to do latent space reasoning, they will probably need a separate model to interpret the intermediate activations no?

I know for some types of ML analysis, a separate model is already used to analyze the weights.

Re: GPT-6 Astra

#115

Post by galaktb »

Cu3PO42 wrote: post #56

Just two days ago, a preprint by Julia Stadlmann went up on arXiv [0] improving the prime gap from 246 to 240. Now OpenAI announces Astra has shown a gap of 186 [1]. That must really blow. [0] https://arxiv.org/abs/2608.31126 [1] https://cdn.openai.com/pdf/51126fac-1b68-4128-9666-c908bcc16...

I think this builds straight upon her method, which she said could be improved herself so...

Re: GPT-6 Astra

#117

Post by MASNeo »

Does anyone feel like everyone chasing the release of Anthropics Fabel 5.1 in a Mad Rush(tm)? In this situation it feels like tuning to benchmarks and other marketing devices feels like trusting Meta in mental health protection of users…

Re: GPT-6 Astra

#118

Post by blargey »

isoprophlex wrote: post #66

> We have found that GPT-6 Astra is more capable of controlling its own CoT than GPT 5.6-Sol, and less likely to include incriminating information in its CoT. In adversarial settings (where we push the model to evade our monitors) we find that the model is able to remain undetected when strategically underperforming in evaluations (sandbagging) and can sometimes evade our internal monitors when asked to perform certa…

"OpenAI is pleased to announce our new model scores 85% on CreateTormentNexusBench - a >60% lead over our leading competitors!"

Did someone get their "AI safety no-no list" and "Frontier features bingo card" mixed up, or did they just stop being able to tell the difference?

Re: GPT-6 Astra

#119

Post by janalsncm »

Legend2440 wrote: post #101
aliljet wrote: post #89

The ARC-AGI-3 score is ridiculously high. Is this benchmaxxing or something way different? It's really hard to discern how we're approaching breakthroughs...

They explain why here: https://openai.com/index/how-two-settings-tripled-our-arc-ag... TL;DR all the other models are being crippled by limitations of their harness. >First, we noticed that after each game action, all private reasoning was discarded. This meant that with each action, GPT‑5.6 Sol was asked to figure out the game anew, unable to remember its past thinking. The model could still see a record of past mov…

Ok so the correct comparison would be to fix the harness on the old model and re-compare. Now they are comparing a new model to an old crippled one.
Post reply on HN