last week, Meituan officially released LongCat 2.0, an open-source model that performs better than Gemini 3.1 Pro on a number of coding benchmarks.
but there was two surprises—first, the model, which performs similarly to DeepSeek V4, had actually been available under a
DAILY SITUATION RECAP (via @theojaffee):
Chinese AI labs are refocusing on domestic chips. Chinese companies currently rely on a mix of smuggled Nvidia chips, legal export-controlled versions of American chips like Nvidia’s H20 and A800 and AMD’s MI308, and Chinese chips like
SITUATION EXPLAINED: Why is software the most important part of the AI hardware stack?
We asked @AnushElangovan, VP of AI software at @AMD
"You got the data center, you got desktop, you got laptops, you got FPGAs. And in time, we have multiple generations of each. So if you map
SITUATION EXPLAINED: A Forethought Institute essay argues the behavioral targets set for today's LLMs will shape future, far more powerful ones.
• The thesis: model specs written for present LLMs will have a large influence on the behavior of future, more powerful LLMs by
AI model specs are usually aimed at shaping the behaviors of present and near-future models. But what if current model behaviors transfer into future models by default?
A new piece looks at possible sources of "inertia" that could lock in today's AI behavioral targets long after
SITUATION EXPLAINED: Why is Apple's seemingly slow AI strategy actually the right one?
@willccbb, Head of Applied Research at @PrimeIntellect
"In my mind, Apple's strategy makes a ton of sense. Apple is able to kind of be slow on adopting use cases because they're kind of
SITUATION EXPLAINED: What is the biggest unsolved problem in RL?
We asked @willccbb, Head of Applied Research at @PrimeIntellect
"There's kinda like two big buckets. One is the environments and data side. One's the algorithm side."
"The environment side is still a very open
SITUATION EXPLAINED: Why is focus the key to success for Frontier Labs?
We asked @willccbb, Head of Applied Research at @PrimeIntellect
"Part of why Anthropic has thrived is because they're not huge. They're not small, but they're not huge either. But also they have a very
SITUATION EXPLAINED: Why is Illinois SB 315 the first state AI safety law with mandatory third-party audits when California and New York both tried and failed first?
We asked @Thomas_Woodside, Co-founder at Secure AI Project.
"Every year AI companies are gonna have to have a
It was a pleasure joining @MTSlive to discuss Pax Silica and the work we are doing at @StateDept to strengthen critical supply chains and expand opportunities for American businesses. Economic security is national security – and we’re putting America’s interests first.
SITUATION EXPLAINED: Why does every current AI safety law have a fundamental gap?
We asked @Thomas_Woodside, Co-founder at Secure AI Project.
"So far we have transparency and auditing. So that means that if a company wants to be transparent about the fact that they're not doing
SITUATION DETECTED: Meta has released Muse Image, its first in-house AI image generation model.
The model will be free via Meta AI, WhatsApp, and Instagram, and will power ad-creation tools in Advantage Plus.
