2026年AI在并购交易撮合中的应用：基准研究报告

How AI is outpacing firms’ ability to control it— and what leaders must do next PRODUCED WITH Contents AI across the deallife cycle06 AI security,governance and risk08 Managing thechange09 Trust, confidenceand human-in-the-loop07 AI tools and the2030 dealmaker10 Contributors Lúcia SoaresChief Information Officerand Head of TechnologyTransformationThe Carlyle Group Amias GeretyPartner, Head of U.S.QED Investors John StecherChief Technology OfficerBlackstone Hari GopalkrishnanChief Technology andInformation OfficerBank of America Byron VielehrChief Operating OfficerApollo Foreword Ken BiscontiSenior Vice President, Co-General ManagerSS&C Intralinks The past twelve months have settled adebate that dominated conversationsamong mergers and acquisitions (M&A)dealmakers over the past two years.Artificial intelligence (AI) is no longerin the experimental phase — it is nowembedded in the deal process. Thecompetitive advantage lies in howeffectively firms deploy it. Nearly halfof the professionals surveyed for thisreport describe AI as fully integratedacross most stages of their dealprocess, with a further four in tenpartially integrated. Equally, senior executives arestatistically more likely to consider AIfully integrated than associates andanalysts, hinting at a potential split inthe overall experience with AI: Seniorexecutives may expect greater digitaltransformations to have occurred, whilethose in more operational roles have adifferent perception altogether. If the adoption question has beenanswered, a more complex set ofquestions has taken its place. Ourresearch reveals that dealmaking is aprofession that is managing growingstacks of AI tools (typically three tofive per team, and in some cases manymore), while simultaneously confrontingsecurity exposures, governance gapsand an organizational resistance that This, however, is caveated by how just38 percent describe AI tools as “wellintegrated” with core platforms usedfor dealmaking, suggesting these AIcapabilities are largely off-platform. This year’sAI in M&A Dealmaking:A Benchmark Study, which is theproduct of a survey of 400 senior dealprofessionals across five organizationtypes and multiple geographies,reveals critical insights into a marketat an inflection point. The technologyis demonstrably ahead of theorganizational structures designed togovern it. The report examines where AI isdelivering measurable value, wheretrust is being extended and where it isbeing withheld, and what dealmakersexpect the profession to look likeby 2030. Critically, it documents thesecurity and governance landscapewith a level of specificity that webelieve will be valuable to anyorganization seeking to calibrate how AI is being applied specificallyacross dealmaking. is intensifying at precisely the momentwhen it might be expected to recede. Furthermore, four in five firmsexperienced an AI-related securityincident in the past year and, likely asa direct consequence, nearly six in tenreport that senior-level pushback hasincreased. These are not the hallmarksof a technology transition that has beenfully absorbed. We are grateful to the 400professionals whose candor andinsight made this report possible. It isour hope that the findings that followprove both informative and useful asyou navigate the next phase of AI-enabled dealmaking. AI in the deal life cycle Valuation andmodeling Historical dealanalytics Due diligence Sourcing andscreening Deal execution Deal marketing Dealmaking teams canexpect time savings ofmore than 11 percentof time during the dealexecution stage, usingAI to handle Q&A-related tasks and detectany red flags. AI is applied mainlyduring early-stagediligence andconfirmatory Q&A,with additional usein financial analysis,anomaly detection anddocument review. Sixty-five percent ofrespondents useAI to produce executivesummaries, with60 percent usingit to developmarketing and dealpositioning strategies. Analysts and associatesare more likely to beusing AI regularly forvaluation and modeling,however 12 percentof partners andmanaging directors(MDs) avoid using AI forvaluation tasks. Most commonly usedto identify patternswithin successful deals,just three percent ofdealmakers said theydo not use AI forhistorical analysis. Teams can expect tosave 21 to 30 percentof their time throughAI, using it for financial/market signal extractionand company discovery. Key findings 80% 33% 49% Eighty percent of dealmakersexperienced AI-related security andaccuracy incidents in the past 12 months. AI has crossed the adoption tipping point,but the complexity is only beginning. AI is delivering measurable time savingsacross the deal life cycle. Nearly half (49 percent) of dealmakers report that AI is fullyintegrated across most deal stages, with a further 41 percentreporting partial integration. Just one in ten remain in pilot orexperimental projects. The typical deal team now manages threeto five AI tools simultaneously (52 percent), with almost