2025年第三季度AI公共报表和估值指南（英）2025

EMERGING TECH RESEARCH AI Public Comp Sheetand Valuation Guide Key takeaways PitchBook Data, Inc. Nizar TarhuniExecutive Vice President of Researchand Market Intelligence Paul CondraGlobal Head of Private Markets Research Stock returns James UlanDirector of EmergingTechnology Research AI core conglomerates led by Chinese firms: Public markets in Q3 favored scaled AI infrastructure and monetization leaders, with Chinesefirms posting the strongest YTD gains among AI core conglomerates. Alibaba led with a 113% YTD return, supported by its multiyearRMB 380 billion (about $53 billion) commitment to AI and cloud infrastructure. The company’s capital intensity and integration acrosse-commerce, logistics, and cloud services reinforced investor confidence in its ability to monetize inference workloads across its platform. Institutional Research Group Analysis Dimitri ZabelinSenior Research Analyst,AI and Cybersecruitydimitri.zabelin@pitchbook.com Baidu rose 61% YTD on momentum in its ERNIE model series and steady growth in AI cloud, where vertical adoption in finance andautomotive has gained traction. Tencent advanced 58%, benefiting from AI feature deployment across social, gaming, and advertising.Oracle gained 69% YTD as OCI demand accelerated and AI-driven workloads expanded its enterprise footprint. pbinstitutionalresearch@pitchbook.com Published on October 28, 2025 On a quarterly basis, Alibaba and Baidu again led, while Tesla climbed 41% on record Q3 deliveries, guidance revisions, and rising investorfocus on autonomy and robotics initiatives. The outperformance of Chinese conglomerates reflected a broader re-rating of Chinesetechnology equities after years of discounting, aided by renewed policy support and heavy state-aligned AI investment. Key takeaways2Stock returns4Revenue6EBITDA8 AI pure plays extend momentum: Performance among AI core pure plays remained exceptional. Palantir rose 141% YTD as its AI Platformscaled across government and enterprise clients, driven by a record quarter exceeding US $1 billion in revenue and expanding multiyearcontracts. Tempus gained 139%, propelled by rapid top-line growth in precision medicine and the integration of newly acquired genomicassets. AppLovin advanced 122% YTD and led the quarter with a 114% gain, reflecting sustained outperformance of its AXON ad-intelligence engine and repeated guidance raises. MongoDB climbed 52% in Q3 as stronger Atlas consumption and stabilizing developer-spend patterns restored investor confidence. The pure-play rally underscored market preference for proven AI monetization, recurringrevenue models, and operating leverage. Investors favored scaled operators translating AI adoption into tangible profitability rather thanspeculative pipeline exposure. PitchBook clients can accessthe full Excel data packfor thisreport via the Details tab in thedocument viewer. AI semiconductors sustain leadership: Semiconductor stocks again outperformed amid elevated infrastructure spending and persistentdemand for high-performance compute. SK hynix led with a 110% YTD gain, driven by record high-bandwidth memory sales and expandingmargins from advanced memory products. Cambricon surged 106% YTD and 145% in Q3 as China’s domestic accelerator build-out andeasing supply constraints boosted sentiment. Micron returned 99% YTD and 37% in Q3, benefiting from a recovery in DRAM and NANDpricing and normalization of datacenter inventory. Astera Labs followed with a 121% quarterly gain on strong demand for server connectivityand interconnect solutions. Across sectors, Q3 performance rewarded companies demonstrating both AI infrastructure control and clearcommercialization. Capital rotation favored names combining credible execution with long-term investment visibility, while policy andcapital-expenditure momentum—particularly in China—amplified valuation expansion across the AI ecosystem. Valuations Valuations have moved unevenly across the AI ecosystem in Q3 2025. AI core conglomerates saw a modest re-rating, with weightedEV/TTM revenue multiples rising from 7× to 8×, reflecting steady top-line growth and sustained investor confidence in scaled AIinfrastructure and monetization. AI core pure plays advanced from 22× to 37×, extending their premium as investors continued to favorhigh-growth, software-driven platforms with strong recurring revenue models. In contrast, AI vertical conglomerates compressed from 10× to 8×, reflecting slower enterprise adoption and lengthening deploymentcycles for large, multiproduct suites. AI vertical pure plays inched up from 14× to 16×, supported by improving profitability and a clearerpath to monetization for specialized AI applications. AI semiconductors increased modestly from 18× to 19×. Rapid shares increases in Q3 were offset by multiple quarters of better-than-expected revenue, higher pricing, and increased CapEX-driven expansion outlooks. AI autonomous machines held increased8x to 9x, consistent with temp