Alibaba (Qwen)

Harvey, the OpenAI-backed legal AI company, published its first post-trained model on 20 August. Harvey Tenet is a Kimi K3 base, post-trained with a rank-64 LoRA over the full network on a dataset of about 1,750 tasks. It completes almost twice as many held-out tasks on the Legal Agent Benchmark as base Kimi K3, and places second on that benchmark overall. Harvey ran ablations on candidate judge models against heavier frontier models and settled on Kimi 2.6 as the grader. Its M&A diligence work uses a GLM-5.2 orchestrator, taking rubric criteria pass rate from a 43.8 percent best baseline to 60.1 percent after self-distillation post-training.

The Central Cyberspace Affairs Commission published a 2026 to 2030 action plan for cyber-information enterprises on 21 August: seven actions and twenty-one measures. Measure 1 orders a continuously updated national database of these firms, with classification and grading standards sorting them into three named tiers. Measure 19 tells regulators to speed up domestic listing review and to open non-IPO exits through S-funds and M&A funds, and tells banks to treat patents, trademarks and data as credit-assessment inputs. Measure 16 orders administrative inspections regulated so as to minimise disruption to normal operations, and Measure 5 names high-end AI chips, high-performance training clusters, world models, agents and embodied intelligence.

Chinese cloud vendors are reselling competitors’ models at roughly 30 percent of list price, below their own cost, because in AI coding the customer names the model. Leiphone reported the discount ladder on 21 August from named cloud sales sources: 60 to 80 percent of list six months ago, 48 percent now common, roughly 30 percent at the low end. The same day, Approaching.AI vice-president Guan Jiawei told the South China Morning Post that domestic chips can only serve the low-quality inference tier, the one with weak demand and weak monetisation, and that high-quality tokens, especially coding, still depend on Nvidia. iFlytek said the same week that a flagship model trained entirely on domestic compute would reach the domestic first tier on code capability.

Forty-eight central state-owned enterprises appeared together at the World Robot Conference in Beijing on 19 August, showing twelve application scenarios under a joint MIIT and SASAC notice that orders each central enterprise to open at least ten real work stations and each designated province at least twenty, targeting more than a hundred high-value scenarios by the end of 2026. The next morning on the same stage Unitree founder Wang Xingxing said robots are not yet efficient or general enough to be rolled out into factories at scale, and put embodied AI’s ChatGPT moment two to three years away at best. His stock fell 16.05 percent that day.

DeepSeek open-sourced Harness, an MIT-licensed agent framework, on 13 August. Within four days its repository had passed both anthropics/claude-code and openai/codex on GitHub stars, 148,734 against 141,759 and 106,470, while Anthropic’s repository carries no open-source license at all. The caveats are in the same API response: Harness is a pre-1.0 release candidate with its public issue tracker switched off, and stars measure attention rather than use.

Alibaba’s Qwen now accounts for more than half of every open-source AI model download on earth and has passed a billion downloads on Hugging Face, displacing Meta’s Llama, per SCMP and Pandaily, though those share figures trace to a single report. The lead is firm enough that Alibaba plans to start charging heavy users of its next open model, and in the same week Meta and Nvidia both re-entered the open-weight market they had backed away from, Meta with Muse Glimmer and Nvidia with Nemotron 3.5 Lightning, explicitly to catch the Chinese labs now setting the open-source default.