Both versions are aligned block by block, in reading order: title, the essentials, then paragraph by paragraph. Where the translation merged or split a paragraph, the matching cell stays empty — we never pair two passages by guesswork.
10万元、日本円で約235万円。AIにこれを渡して、1年間ネットショップを回させる。中国Alibabaの Qwenチームと香港科技大学（HKUST）の研究者らがまとめた論文「E-Commerce Bench」は、そんな長期の模擬経営でAIモデルの自律的なビジネス運用能力を測るベンチマーク（同じ条件で複数のモデルを比べる共通の試験）を提示した。論文はarXivに公開された予備稿（プレプリント）である。
100,000 yuan, about 2.35 million yen. The AI is given this sum and told to run an online shop for a year. A paper titled “E-Commerce Bench,” compiled by researchers from Alibaba's Qwen team and the Hong Kong University of Science and Technology (HKUST), presents a benchmark (a common test that compares multiple models under the same conditions) that measures the autonomous business capabilities of AI models through this kind of long-term simulated operation. The paper is a preprint posted on arXiv.
舞台は、実際のECサイト、淘宝（タオバオ）と天猫（Tモール）のデータに基づく市場モデルだ。6886点の商品と576社のサプライヤーが存在し、AIは最大4店舗を同時に運営する。市場調査から価格交渉、仕入れ、価格設定、発送、返品処理、資金管理まで、18種類のツールを使って業務を自律的に回す。サプライヤーの約4分の1は、相場より高く売りつける、注文数をごまかす、返品率の跳ね上がる粗悪品を送る、といった詐欺業者として設定されている。
The setting is a market model based on data from real e-commerce platforms, Taobao and Tmall. The market contains 6,886 products and 576 suppliers, and the AI operates up to four stores at the same time. Using 18 types of tools, it autonomously handles everything from market research to price negotiation, procurement, pricing, shipping, returns processing and cash management. About one-quarter of the suppliers are set up as scammers that overcharge relative to market prices, falsify order numbers or ship low-quality goods with soaring return rates.
資金の流れにも罠がある。仕入れ代や運営費は銀行口座から即座に引き落とされる一方、売上金は発送から9日間、エスクロー（取引の安全のために第三者が代金を預かる仕組み）口座に留め置かれてからウォレットに入る。そのため、黒字でも計画的に銀行口座へ出金しなければ、銀行口座のマイナスが10日連続で続き、破産となる。さらに、古い会話履歴は一定量を超えると自動で消えるため、AIは最大20件のメモリを使って取引の情報を覚えておかなければならない。
The cash flow also has traps. Procurement and operating costs are debited immediately from the bank account, while sales proceeds are held for nine days after shipment in an escrow account (a system in which a third party holds payment to secure the transaction) before they reach the wallet. Therefore, even a profitable store will go bankrupt if it does not plan withdrawals to its bank account, as the bank balance remains negative for 10 consecutive days. In addition, old conversation history is automatically deleted once it exceeds a certain amount, so the AI must use up to 20 memory entries to keep track of transaction information.
結果で首位に立ったのは「GPT-5.6 Sol」で、初期資金を平均約14.3倍（約143万元）に増やした。ただし詐欺業者への支出割合は18モデル中16位と悪かった。2位の「Fable5」（約8倍）はGPT-5.6 Solの半分以下のツール呼び出し回数で手堅く稼ぎ、1手あたりの利益効率では1位だった。「Claude Opus 4.7」は値引き交渉と詐欺回避の両方で全モデル中トップの緻密さを見せたが、取引規模を広げられず総資産は8位にとどまった。オープンウェイトモデルの最高は、アリババの「Qwen3.8-Max-Preview」（約4.2倍、全体5位）で、再仕入れの際に過去の最安値を基準に値引きを引き出す長期学習能力が全18モデル中トップだった。
The top performer was “GPT-5.6 Sol,” which grew its initial funds by about 14.3 times (about 1.43 million yuan) on average. However, it ranked 16th of 18 models for the share of spending directed to scam suppliers. “Fable5” (about eight times), in second place, earned steadily with less than half the number of tool calls used by GPT-5.6 Sol, and ranked first in profit efficiency per action. “Claude Opus 4.7” showed the most meticulous performance among all models in both price negotiation and scam avoidance, but it failed to expand its trading volume, leaving its total assets in eighth place. The best open-weight model was Alibaba's “Qwen3.8-Max-Preview” (about 4.2 times, fifth overall), which had the top long-term learning capacity among all 18 models, using the lowest past purchase price as a benchmark to win discounts when restocking.
この結果が効くのは、AIに業務を任せたい事業者と、モデルを選ぶ開発者である。資産を大きく増やすモデルと、詐欺や破産を避けるモデルが同じとは限らないことを、論文は数字で示した。全90回の試行のうち10回が途中で破産しており、内訳はGPT-5.5、Claude Opus 4.6、Gemini 3.1 Proが5回中2回、Qwen3.5-Plusが5回中4回だった。ただし各モデルのテストは5回ずつで、市場も模擬のものだ。
These results matter to businesses that want to hand over operations to AI and to developers choosing models. The paper shows with numbers that the models that grow assets the most are not necessarily the ones that avoid scams and bankruptcy. Of the 90 trials in total, 10 ended in bankruptcy partway through. By model, GPT-5.5, Claude Opus 4.6 and Gemini 3.1 Pro went bankrupt in two of five runs each, and Qwen3.5-Plus in four of five. However, each model was tested only five times, and the market is simulated.