Cities, Transport and AI. Exploring practical developments and perspectives on artificial intelligence. https://kimkj.com/podcast/urban-ai-en en Kyungjin Kim Kyungjin Kimkimkj008@gmail.com episodic false A compilation for 31 October 2026 built from material provided by attorney Kyungjin Kim, focused on introducing and explaining the original sources. 1. On-board AI vision for Spanish waste-collection trucks: MOBA MAWIS.AI VIEW 2. Taiwan Water Corporation finds 4,003 underground leaks with acoustic AI and big data 3. ETRI and Daegu: a 2025 digital-twin water-network control pilot 4. A Midwest suburban water utility pinpoints a two-year hidden leak with acoustic AI Each segment was generated separately as a Brief from a single selected source and then joined in order. The programme introduction appears only in the first segment. Claims, figures and assessments belong to the cited sources, not to attorney Kim. Audio was AI-generated. The closing provider notice at the end is a Korean AI voice line. kimkj-urban-ai-en-20261031 Sat, 31 Oct 2026 10:32:11 +0900 341 false An English compilation of four individually selected NotebookLM Briefs, based on material provided by attorney Kyungjin Kim. Actual duration: about 5 minutes 2 seconds. 1. Jeju, reported by Jeju Traffic Magazine on 9 July 2026: the autonomous police and Korea Road Traffic Authority used camera/plate/flow observations, VISSIM timing-plan simulations and field controllers on a 13.4 km Aejo-ro corridor. The May 26–June 19 results include 31.3% less delay and 14.1% higher speed in the driving survey versus 4.1% higher speed in the system data. The source gives no causal explanation for that difference. KRW 30.9 billion is an estimated annual social/economic benefit, not proven revenue or ROI. Two more sections are a plan. 2. Hong Kong Transport Department, news.gov.hk, 20 August 2026: the Q1 2027 Causeway Bay pilot and projected 15–20-second pedestrian wait reduction are future plans, not achieved operating results. The separate 50-junction rollout is also planned. The licence-booking result—2 hours to under 45 minutes after late-June 2026 AI bot controls—is a separately reported outcome. 3. Vitoria-Gasteiz, Europa Press, 26 December 2023: a historical Project Green Light adoption plan. AI uses aggregated Maps/Waze data to recommend timing; officials apply it. The 30% and 10% figures are the reported global field-test figures, not a local Vitoria measurement or a 2026 update. Pedestrian buttons, bus priority and phased selection of intersections remain relevant conditions. 4. Delhi, Economic Times/PTI, 24 October 2026: the government announced an AI-enabled command centre, construction-dust cameras/GPS measures, and 13 stations bringing the network to 60. No quantified emission-reduction result is asserted. The original CREA analyst's assessment is retained: monitoring visible dust alone is insufficient for secondary PM2.5, about 25% of winter PM2.5, formed from SO2, NOx and ammonia. Neutral comparisons explain the provided methods. Unsupported gap causes, profit/ROI claims, imagined enforcement questions and broad claims that AI can never address gases were omitted. The existing GUID and original publication date are retained. This existing r2 revision predates the new-outro policy and is not retrofitted with it. AI-generated and edited; no audio-speed changes or length-filling silence. kimkj-urban-ai-en-20261029-trial Thu, 29 Oct 2026 14:20:15 +0000 302 false