Of course that's your contention. You're a first-time SaaS bear. You just got finished listening to some podcast, Dario on Dwarkesh, probably. Now you think it’s the end of white collar work and seat-based pricing is screwed. You're gonna be convinced of that til tomorrow when

This means there's actually a huge productivity boom underway, btw. It's AI.

How efficient is MiniMax M2.5? We benchmarked on 8xH200 TEP8 with vLLM . At a reasonable 10-25s TTFT, M2.5 is able to sustain ~2500 tok/s/GPU throughput. For decode, it's still possible to reach ~20 tok/s/GPU throughput at a strict 20 tok/s/user interactivity with 10K+

NIK >Anthropic partners with literal Department of War >Is surprised its AI is used in War applications

Like I've been saying for many years

in the UK, bureaucrats enroll your kids in school and keep the gov money the UK hates its children

As someone works in human rights, I’ve never seen anything like the Epstein files in my 15-year career Raped Cannibalized Trafficked Filmed Terrorized Tortured Murdered 13, 14, 15 year children ZERO arrests I don't understand how we're not having a global revolution right now

la mala onda de la gente cuando emprendes

.Dario Amodei gets the CEO job: "Culture is the only thing that scales at this speed. Dario spends 30-40% of his time on culture. Not products. Not strategy. Culture. Every two weeks he gets up in front of the entire company for an hour: a ritual called the DVQ (Dario Vision

4chan: today's news three years ago.

OpenAI and Anthropic are earning $11-14B a year while burning $100B's and being unprofitable, while Apple just licensed Gemini for $1B/year and stands to increase $20-80B on hardware sales if they can get their AI products right. Then, they'll release M5 chips and customers will

Moonshot AI's Kimi K2.5 is the #2 open weights model, and is now available across 8 providers, each with differing price and performance Kimi K2.5 from Kimi.ai is one of the leading open weights models, scoring 47 on the Artificial Analysis Intelligence Index, and its

HOLY F**KING SH*T

Gemini Ultra does make any sense right now. + the billing is broken. they charge on one day and renew on another, you don't know how much you have left + Gemini gets mogged by open source. It is severely unstable + Veo 3 gets beaten by Seedance 2.0 + The 30TB cloud...for what?

Alibaba's new Qwen3.5-397B-A17B is the #3 open weights model in the Artificial Analysis Intelligence Index - a significant upgrade from Qwen3-235B-A22B-2507, and achieved with fewer active parameters than leading peers Qwen3.5-397B-A17B is the first model released by Alibaba

The birth rate is consistently crashing faster than any Overton window for our earlier computer models. Our earlier computer models had massive standard deviations so that’s basically the equivalent of a comet striking statistically, which means something is very very wrong. The

Lmao

Grok 4.20 is BASED. The only AI that doesn’t equivocate when asked if America is on stolen land. The others are weak sauce.

I hate that Microsoft might be vibecoding Windows, but it's inevitable microsoft laid off everyone who knows how c++ works so now they just prompt gpt 5 to fix the codebase. 30% of windows is written by ai. that is why your printer drivers were deleted to make room for 4gb of

Presenting the GLM-5 Technical Report! arxiv.org/abs/2602.15763 After the launch of GLM-5, we’re pulling back the curtain on how it was built. Key innovations include: - DSA Adoption: Significantly reduces training and inference costs while preserving long-context fidelity -