https://microsoft.ai/wp-content/uploads/2026/06/main_2026060...
Launching seven new MAI models: https://microsoft.ai/news/building-a-hillclimbing-machine-la...
MAI-Thinking-1
microsoft.ai
MAI-Thinking-1
1–10 of 90 posts
Re: MAI-Thinking-1
#2Absolutely disgusting scroll jacking, even when "Accessibility mode" is turned on
Re: MAI-Thinking-1
#3Looks like the OAI divergence is finally taking place. Seems like the comparisons are mainly with Opus 4.6 and GPT 5.4 though. Still, exciting to see a new frontier player.
Re: MAI-Thinking-1
#4Absolutely disgusting scroll jacking, even when "Accessibility mode" is turned on
I'm sure most of us agree, but:
"Please don't complain about tangential annoyances—e.g. article or website formats, name collisions, or back-button breakage. They're too common to be interesting."
Re: MAI-Thinking-1
#5> Second, clean data. MAI-Thinking-1 was trained on clean and appropriately licensed data, with AI-generated content excluded from pre-training. This matters for quality, provenance, and control. If we cannot account for what shaped a model, we cannot fully understand its behavior or credibly improve it.
Shots fired?
It would be interesting to see how far "clean data" can go on the scaling laws.
Re: MAI-Thinking-1
#6> Second, clean data. MAI-Thinking-1 was trained on clean and appropriately licensed data, with AI-generated content excluded from pre-training. This matters for quality, provenance, and control. If we cannot account for what shaped a model, we cannot fully understand its behavior or credibly improve it. Shots fired? It would be interesting to see how far "clean data" can go on the scaling laws.
I'm interested how much "Clean Data" is synthetic data from "unclean" models...
Re: MAI-Thinking-1
#7> Second, clean data. MAI-Thinking-1 was trained on clean and appropriately licensed data, with AI-generated content excluded from pre-training. This matters for quality, provenance, and control. If we cannot account for what shaped a model, we cannot fully understand its behavior or credibly improve it. Shots fired? It would be interesting to see how far "clean data" can go on the scaling laws.
I'm interested how much "Clean Data" is synthetic data from "unclean" models...
“ We trained it from the ground up on enterprise grade, clean and commercially licensed data, without distillation from third-party models.”
Re: MAI-Thinking-1
#87 modes launched. 5 models in the dropdown. Only 4 actually usable :(
About time Microsoft joined the fray. After the OpenAI divorce, it really looked like Microsoft was going to become another Uber.
Re: MAI-Thinking-1
#9We need to see DeepSWE scores. SWE Bench Pro is junk.
Re: MAI-Thinking-1
#107 modes launched. 5 models in the dropdown. Only 4 actually usable :( About time Microsoft joined the fray. After the OpenAI divorce, it really looked like Microsoft was going to become another Uber.
They still own 27% of OpenAI, this IPO will feed them a lot of easy cash.