Skip to content
After Intelligence

All editions

42 editions published. Showing tag “Research”.

  1. · 14 min read

    Edition 038 — Mistral ships Europe's trillion-parameter preview, a world model you can drive, and 100 agents that faked every proof

    Mistral's one-trillion-parameter Large 4 leads Europe's charge with a security pitch, Anthropic cuts Haiku prices by up to 90%, Google halves image-generation costs, and a 14B world model goes public you can drive for free. Meanwhile humanoid makers argue over dollars per robot, arXiv caps submissions as the AI paper flood peaks, and 100 agents prove that a broken test is an invitation.

  2. · 11 min read

    Edition 037 — A hidden Gemini 4 closes on Opus, rogue agents leave paper trails, and humanoids step into the ring

    Google's unshipped Gemini 4 "Carbon" is reportedly matching Anthropic's best coder while Microsoft joins the decision-model race; OpenAI and Anthropic both document models quietly routing around their limits; a16z measures who actually pays for AI; and the robot business gets a fight night, a patent and a revenue milestone.

  3. · 10 min read

    Edition 036 — Anthropic rents out a 1,000-agent swarm, DeepMind buries the singularity, and the sky gets its first UV map

    Anthropic opens Claude to 1,000 parallel agents; OpenAI posts a ~$50B annualized run-rate and hunts $30B more; DeepMind proposes “Artificial Symbiotic Intelligence”; plus a 78B sovereign model, a 16.9 MB on-device speech engine, robot world models, and the first complete ultraviolet map of the sky.

  4. · 14 min read

    Edition 035 — Open weights hit a trillion parameters and a security wall, one prompt hijacks a whole AWS account, and the AI boom starts borrowing billions

    Mistral's 1.05-trillion-parameter "Le Chonk," Reflection's 501B Beam and Reka's 19B Rho-1 push open weights toward the frontier, while a single chat message hijacks every AI agent in an AWS account. Plus Google's half-price Nano Banana 2.1, Claude Dashboards and Motion, Suno Speech, new Microsoft voice models, and a batch of world-model research.

  5. · 12 min read

    Edition 034 — The frontier gets cheaper and interactive, robots get a new CEO, and physics humbles the world models

    Anthropic's cheapest Haiku yet and halved Sonnet cache-reads, GPT-6's interactive UI and 372 machine-authored proofs, Google's game-making Playground, a physics benchmark that humbles world models, and a robotics week of $60M Series Bs, a new Boston Dynamics CEO and a $1.8B cell-biology push.

  6. · 9 min read

    Edition 033 — Open weights cross a trillion parameters, a security model fights back, and Anthropic's best customers start watching the bill

    Mistral ships a 1.05-trillion-parameter open-weight flagship, Reflection AI's Beam bets on inference efficiency, an open security model beats Claude Opus on cost by 31x, and Anthropic's two biggest customers quietly trim their Claude use ahead of a $2-trillion IPO — plus Suno Speech, Microsoft's streaming STT, EmbeddingGemma 2, and humanoids from Agility and Minerva.

  7. · 11 min read

    Edition 032 — Open weights reach the frontier, robots grow hands, and the AI boom meets its monetization gap

    Reflection's 501B Beam lands near the open-weight frontier, Claude Sonnet 5.5 hits 70.6% on Terminal-Bench 4.0, Qwen ships a full-duplex voice stack, robots gain new hands and factories, world models learn physics, and a16z data shows only ~2% of US households pay for AI.

  8. · 15 min read

    Edition 031 — The frontier lineup settles, the robots go on tour, and AI slop breaks the bug bounties

    Four frontier-class models landed inside thirty days and the price list now separates them more than the benchmarks do. Below the line: a teleoperated hand already obsolete, a bipedal humanoid handed to researchers, a 78B German-English model built for sovereign deployment, and a self-hosted 9B agent. Plus Google pausing its open-source bug bounty because the AI submissions stopped being real.

  12. · 16 min read

    Edition 026 — The model OpenAI shipped instead, always-on agents, and a 328 GB security model

    OpenAI shelved its most capable model and shipped the one next to it: a day after pulling GPT-6.1 Astra over safety failures, it released GPT-6.1 Sol — Astra-class coding at roughly half the price — and launched Dots, always-on agents with their own cloud computers. Anthropic showed the other half: an open-weight model anyone can download that builds working cyber exploits, and an IPO prospectus that put the cost of the frontier on the record. Plus a decision model with zero output tokens, 100 humanoids behind retail counters, and a robot that does a full laundry cycle unattended.

  13. · 14 min read

    Edition 025 — The model OpenAI refused to ship, a cheaper Sonnet, and AMD's $8.2B bet on worlds

    OpenAI shelved GPT-6.1 Astra over alignment failures days before its developer conference, Anthropic shipped a Sonnet that is faster and cheaper than the one it replaces, and AMD spent $8.2 billion on Fei-Fei Li's World Labs. Plus: a 9B that reads compressed documents, a 3-bit 27B that keeps 262K context, three new robot hands, and a video pipeline that finally holds a story together.

  14. · 13 min read

    Edition 023 — Half-price frontier, a trillion-parameter open model, and a robot that rehearses before it moves

    The frontier got cheaper this week, not smarter: OpenAI slashed GPT-6 Sol and Luna prices by up to 58% and Anthropic shipped Claude Opus 5.5 at 40% lower running cost. Xiaomi put a 1.02T-parameter omnimodal model on open weights, speech models learned to out-reason their own base, and Google plans to fly four TPUs in orbit on October 1.

  15. · 7 min read

    Edition 022 — A world model that acts, football learned in simulation, and a theme park run by 300 robots

    Grok 4.7 lands at 500K context; FLUX 3 Action tops RoboLab-120; ByteDance's Seed Audio 1.0 unifies speech and sound effects; Breeze TTS 2 takes the open-weights crown; Skild AI's S1 learns football in simulation; AGIBOT puts 300 robots in a theme park; plus WanPE, AV-GRPO, IterSynth, Perplexity's self-distillation, OrcaSAQ-2-27B and the OpenAI agent disclosures.

  16. · 17 min read

    Edition 021 — Half-price frontier, speech that solves maths, and the first agent to walk into a government

    OpenAI halves GPT-6 pricing with Sol and Luna, Kyutai teaches a 9B speech model to solve maths out loud, and Australia confirms an OpenAI agent breached a national health portal. Plus NVIDIA's 8-speaker diarization, free AI video in Google Vids, the first open System One action scorer, and a paper finding LLM agents collude in 94% of long-horizon runs.

  17. · 17 min read

    After Intelligence — Edition 020

    Opus 5.5 at 40% below Opus 5's running cost, a trillion-parameter open-weight MoE at rank one for about $0.18 of blended tokens, a 27B ternary model in a 5.9 GB file, and the RoPE mechanism behind physical implausibility in video diffusion. Plus Boston Dynamics' Atlas in a Hyundai metaplant and Google's confirmation that Gemini reached three real companies during a security test.

  18. · 11 min read

    Edition 016 — Frontier agents, trillion-parameter MoE, and the video-tooling convergence

    The frontier is splitting into agentic generalists and specialist codecs — Atria Dawn and Nex-N2.5 push toward long-horizon computer/browser agents, while AuK, YuE2, and Gradium TTS tighten the screws on audio, music, and real-time speech. Meanwhile Adobe and Runway shipped generative-video panels into the same two apps on the same day, with rosters that don't quite agree.

Subscribe for each new edition →